WO2005075156A1 - 移動ロボットの歩容生成装置 - Google Patents
移動ロボットの歩容生成装置 Download PDFInfo
- Publication number
- WO2005075156A1 WO2005075156A1 PCT/JP2005/001693 JP2005001693W WO2005075156A1 WO 2005075156 A1 WO2005075156 A1 WO 2005075156A1 JP 2005001693 W JP2005001693 W JP 2005001693W WO 2005075156 A1 WO2005075156 A1 WO 2005075156A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- gait
- model
- instantaneous
- arrangement
- posture
- Prior art date
Links
- 230000005021 gait Effects 0.000 title claims abstract description 763
- 230000033001 locomotion Effects 0.000 claims abstract description 466
- 238000012937 correction Methods 0.000 claims abstract description 378
- 238000006243 chemical reaction Methods 0.000 claims abstract description 339
- 230000005484 gravity Effects 0.000 claims description 70
- 230000001133 acceleration Effects 0.000 claims description 47
- 230000008859 change Effects 0.000 claims description 28
- 238000005452 bending Methods 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 12
- 230000002123 temporal effect Effects 0.000 claims description 7
- 230000036544 posture Effects 0.000 description 555
- 238000006073 displacement reaction Methods 0.000 description 508
- 210000002683 foot Anatomy 0.000 description 134
- 238000012545 processing Methods 0.000 description 123
- 210000002414 leg Anatomy 0.000 description 110
- 238000000034 method Methods 0.000 description 102
- 230000008569 process Effects 0.000 description 94
- 239000000047 product Substances 0.000 description 72
- 210000000689 upper leg Anatomy 0.000 description 54
- 210000000629 knee joint Anatomy 0.000 description 27
- 238000010586 diagram Methods 0.000 description 25
- 230000006870 function Effects 0.000 description 21
- 210000004394 hip joint Anatomy 0.000 description 21
- 210000000544 articulatio talocruralis Anatomy 0.000 description 19
- 210000001503 joint Anatomy 0.000 description 16
- 239000013598 vector Substances 0.000 description 14
- 239000002131 composite material Substances 0.000 description 13
- 238000013459 approach Methods 0.000 description 11
- 230000007246 mechanism Effects 0.000 description 11
- 210000002310 elbow joint Anatomy 0.000 description 9
- 230000000694 effects Effects 0.000 description 8
- 230000001965 increasing effect Effects 0.000 description 8
- 238000003384 imaging method Methods 0.000 description 7
- 230000014509 gene expression Effects 0.000 description 6
- 238000013519 translation Methods 0.000 description 6
- 230000009471 action Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 5
- 230000007423 decrease Effects 0.000 description 4
- 239000000463 material Substances 0.000 description 4
- 210000000323 shoulder joint Anatomy 0.000 description 4
- 230000001502 supplementing effect Effects 0.000 description 4
- 210000003857 wrist joint Anatomy 0.000 description 4
- 210000003423 ankle Anatomy 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 101710153181 Protein PYRICULARIA ORYZAE RESISTANCE 21 Proteins 0.000 description 2
- 230000010391 action planning Effects 0.000 description 2
- 230000037396 body weight Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 210000003127 knee Anatomy 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 230000006641 stabilisation Effects 0.000 description 2
- 238000011105 stabilization Methods 0.000 description 2
- 241001481789 Rupicapra Species 0.000 description 1
- 239000000853 adhesive Substances 0.000 description 1
- 230000001070 adhesive effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000001010 compromised effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 239000013013 elastic material Substances 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 230000001771 impaired effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000001568 sexual effect Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
- 210000000707 wrist Anatomy 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J5/00—Manipulators mounted on wheels or on carriages
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B62—LAND VEHICLES FOR TRAVELLING OTHERWISE THAN ON RAILS
- B62D—MOTOR VEHICLES; TRAILERS
- B62D57/00—Vehicles characterised by having other propulsion or other ground- engaging means than wheels or endless track, alone or in addition to wheels or endless track
- B62D57/02—Vehicles characterised by having other propulsion or other ground- engaging means than wheels or endless track, alone or in addition to wheels or endless track with ground-engaging propulsion means, e.g. walking members
- B62D57/032—Vehicles characterised by having other propulsion or other ground- engaging means than wheels or endless track, alone or in addition to wheels or endless track with ground-engaging propulsion means, e.g. walking members with alternately or sequentially lifted supporting base and legs; with alternately or sequentially lifted feet or skid
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J13/00—Controls for manipulators
Definitions
- the present invention relates to an apparatus for generating a desired gait of a mobile robot such as a bipedal mobile robot.
- Patent Document 1 JP-A-2002-326173
- Patent Document 2 PCT International Publication WOZ03Z057427 ZA1
- the technology found in these documents uses the first dynamic model that represents the relationship between the motion of the robot (the position and posture of each part) and the floor reaction force, and uses the first dynamic model.
- the robot's goal is to satisfy the required dynamic equilibrium conditions (conditions such that the translational force component of the floor reaction force becomes the target value, and the floor reaction force moment around a certain point becomes the target value).
- An instantaneous desired gait consisting of the instantaneous value of the movement (instantaneous target movement) and the instantaneous value of the desired floor reaction force (instant desired floor reaction force) is sequentially created. Then, the instantaneous desired gait is input to the second dynamic model, and a part of the instantaneous desired movement (the desired body position / posture, the desired moment around the desired ZMP, etc.) is corrected, so that the final desired gait is obtained. In such a way that a typical instantaneous desired gait is generated in time series.
- a highly linear model is generally used as the first kinetic model.
- a gait that leads to a normal gait which is a virtual periodic gait, or a gait that approaches asymptotically (stable motion of the robot Gaits that can be continuously performed can be efficiently created in a short time.
- a dynamic model with high linearity generally tends to have relatively low dynamic accuracy in various operations of a robot.
- the dynamics of the robot on the dynamic model is likely to have an error with respect to the actual dynamics of the real robot. For this reason, if the instantaneous target gait created using the first dynamic model is applied to the real robot as it is and the real robot is operated, it is guaranteed on the first dynamic model. Movement The mechanical equilibrium condition is not satisfied on the real robot, and the operation of the real robot tends to be unstable.
- the first dynamic model tends to have low dynamic accuracy as described above, so that a dynamic error is relatively large depending on a gait to be generated. May be.
- the above error tends to be large.
- a three-mass dynamic model in which the upper body of a bipedal mobile robot has one corresponding mass point near the tip of each leg, or a robot with a mass point only in the upper body of the robot 1
- the dynamic model of the mass is used as the first dynamic model, particularly when the knee joint of each leg is bent relatively quickly, the influence of the change in the inertial force accompanying the movement is required.
- the dynamic error becomes relatively large.
- the instantaneous desired gait created using the first dynamics model may be excessively inappropriate for securing the continuous stability of the robot.
- the correction is not performed properly, and the corrected instantaneous target gait has a stability margin.
- the robot may be too low or may not be able to maintain the stability of the robot, and may diverge to the object.o
- the inventor of the present application described the instantaneous desired gait motion created using the first dynamic model without using the dynamic model (representing the relationship between the motion and the force). Without using differential equations or integral equations), the position and orientation of a given part are corrected by geometrical arithmetic processing to increase the dynamic accuracy with the instantaneous target floor reaction force (dynamic dynamics).
- a method for reducing errors was previously proposed in Japanese Patent Application No. 2004-5029.
- This technique For example, the body position and body posture of the instantaneous target gait created using the first dynamic model described above are subjected to geometric calculation processing (the value of the instantaneous target floor reaction force, its time-series value, and the body position. (The arithmetic processing not using the differential value of the posture). This method does not require kinetic calculation processing! /, So it is possible to efficiently correct the instantaneous target motion in a short time.
- the motion of the instantaneous target gait is corrected by geometric calculation processing so as to reduce the dynamic error every time the instantaneous target gait occurs.
- the posture of the affected part (such as the upper body) may fluctuate frequently.
- the correction of the instantaneous target movement for improving the dynamic accuracy can be mainly performed by correcting the body position.
- the floor reaction force corresponding to the inertial force to be generated by the corrected body position motion is actually obtained. May not occur.
- the corrected instantaneous target motion is such as to cause a slip of the robot.
- the present invention has been made in view of a powerful background, and is capable of expressing a motion of an instantaneous target gait created using a dynamic model without using a dynamic model (representing the relationship between motion and force). Improve the dynamic accuracy between the instantaneous desired gait and the floor reaction force without using differential equations and integral equations), and minimize the fluctuation of the posture of a predetermined part such as the upper body of the robot. Can be properly corrected so as to achieve both It is an object of the present invention to provide a gait generation device for a mobile robot that can generate a gait that can be made.
- a first invention of a mobile robot gait generator of the present invention provides an instant gait that sequentially generates an instantaneous target gait comprising an instantaneous target motion of the mobile robot and an instantaneous target floor reaction force.
- a gait generator including a gait generating means, a first temporary correction for temporarily determining a position and a posture of a predetermined portion of the mobile robot from the instantaneous target motion, a first temporary correction for determining an instantaneous target motion Motion determining means, and a second temporarily corrected instantaneous target movement obtained by temporarily correcting the position of the predetermined part while maintaining the posture of the predetermined part from the instantaneous target movement the same as the posture in the instantaneous target movement.
- the position and orientation of the predetermined portion in the instantaneous target movement Based on the first provisional corrected instantaneous target movement and the second provisional corrected instantaneous target movement, the position and orientation of the predetermined portion in the instantaneous target movement. Perform true correction And a desired motion correcting means for determining a positive after instantaneous desired motion. Then, the whole or a part of the mobile robot is represented by a model having a plurality of element forces using at least one of a rigid body having inertia and a mass point as an element, and the instantaneous gait generating means is provided.
- the arrangement of each element of the model determined according to a first predetermined geometric constraint condition that defines the relationship between the instantaneous motion of the mobile robot and the arrangement of each element of the model is defined as a first From the first temporary corrected instantaneous target motion determined by the first temporary corrected motion determining means, a predetermined second geometrical shape defining the instantaneous motion of the mobile robot and the arrangement of each element of the model.
- the arrangement of each element of the model determined according to the constraint condition is defined as a second arrangement, and the second geometric constraint is determined from the second temporarily corrected instantaneous target motion determined by the second temporarily corrected motion determining means.
- the first provisional correction motion determining means determines the arrangement of each element of the model between the second arrangement and the first arrangement.
- the translational force component of the resultant force of the inertial force of each element calculated by considering the difference of the acceleration as the acceleration becomes substantially zero, and the moment component generated around the predetermined point at the resultant force becomes substantially the predetermined value.
- the first temporary corrected instantaneous target motion is determined, and the second temporary corrected motion determining means determines the difference between the arrangement of each element of the model between the third arrangement and the first arrangement.
- the second temporarily corrected instantaneous target motion is determined so that the resultant force of the inertial forces of the respective elements calculated by doing so becomes approximately the predetermined value in the moment component generated around the predetermined point.
- the motion correcting means is configured to multiply the attitude of the predetermined portion in the first temporary corrected instantaneous target motion by a predetermined weight wl and the attitude of the predetermined portion in the second temporary corrected instantaneous target motion.
- the sum of the predetermined weight w2 and the sum is determined as the instantaneous target posture of the predetermined portion in the corrected instantaneous target motion, and the sum of the sum is calculated at the position of the predetermined portion in the first temporary corrected instantaneous target motion.
- the sum of a product obtained by multiplying a predetermined weight wl and a position obtained by multiplying a position of the predetermined part in the second temporary corrected instantaneous target motion by a predetermined weight w2 is the predetermined value in the corrected instantaneous target motion. Part of the moment The time target position is determined.
- the “arrangement” of the element of the model includes the “position” of the mass point as the element and the inertia as the element.
- This is a general term for the "posture” (inclination angle) of a rigid body (link) having a.
- a rigid body has a mass and inertia, but for convenience, in the present invention, a rigid body having its mass and inertia has a mass having the mass and located at the center of gravity of the rigid body. It is assumed that it has been decomposed into a rigid body having a mass of 0 and the inertia. This does not lose generality.
- the terms “first arrangement”, “second arrangement”, and “third arrangement” mean a set of arrangements of all elements included in the model.
- the first geometric constraint condition and the second geometric constraint condition are appropriately set, and the elements constituting the model are appropriately set.
- the difference between the second arrangement and the first arrangement (the difference between the arrangement of each element in the second arrangement and the arrangement of each element in the first element) is determined by the first arrangement.
- the instantaneous gait generating means generates the temporary corrected instantaneous target motion (at least one of the position and orientation of each part of the robot determined by the first temporary corrected instantaneous target motion).
- the magnitude (degree,) of the dynamic error between the instantaneous target floor reaction force (the instantaneous target value of at least one of the translational force and moment of the floor reaction force acting on the robot) can be adjusted. It becomes possible.
- the difference between the third arrangement and the first arrangement is calculated by the second temporary correction instantaneous.
- Target exercise above At least one of the position and orientation of each part of the robot for which the second provisional corrected instantaneous target motion has been determined, and / or any of the deviations
- instantaneous desired floor reaction force generated by the instantaneous gait generating means It is possible to correspond to the degree (degree) of the kinetic error between. Supplementally, there is generally a stationary offset in these correspondences.
- the inertial force of each element calculated by considering the difference in the arrangement of each element of the model between the second arrangement and the first arrangement as acceleration.
- the position and the position of the predetermined portion are set such that the translational force component of the resultant force is substantially zero and the moment component generated around the predetermined point is substantially a predetermined value (a certain offset value).
- the first temporary corrected instantaneous target movement is determined assuming that the posture has been temporarily corrected from the instantaneous target movement generated by the instantaneous gait generating means.
- the dynamic accuracy of the first temporary corrected instantaneous target movement is enhanced with respect to both the translational force component and the moment component of the instantaneous target floor reaction force.
- the posture of the predetermined part in the first temporary corrected instantaneous target movement may frequently change.
- the instantaneous gait generating means when the translational force component of the instantaneous desired floor reaction force is not explicitly set, the instantaneous gait generating means generates a dynamic model used for generating a gait.
- the translational force component is regarded as the translational force component of the instantaneous desired floor reaction force.
- the second temporarily corrected instantaneous target motion as a result of correcting the position of the predetermined portion from the instantaneous target motion generated by the instantaneous gait generating means changes the posture of the predetermined portion.
- the dynamic accuracy is improved with respect to the moment component of the instantaneous desired floor reaction force, and at the same time, the posture of the predetermined portion is changed by the instantaneous gait generating means.
- the posture (instantaneous target posture) of the predetermined part in the corrected instantaneous target movement is a predetermined posture of the posture of the predetermined part in the first temporary corrected instantaneous target movement.
- the weighted wl is determined by the sum of the product obtained by multiplying the posture of the predetermined part by the predetermined weight w2 in the second provisional corrected instantaneous target motion.
- the position of the predetermined part (instantaneous target position) is obtained by multiplying the position of the predetermined part in the first temporary correction instantaneous target movement by the predetermined weight wl and the second temporary correction instantaneous target movement. It is determined as the sum of the position of the predetermined part multiplied by a predetermined weight w2.
- the dynamic accuracy between the instantaneous target floor reaction force and the instantaneous target floor reaction force can be suppressed while suppressing excessive fluctuation of the instantaneous target posture of the predetermined portion.
- the corrected instantaneous target motion that can be secured better than the instantaneous target motion generated by the instantaneous gait generating means.
- the first temporary corrected instantaneous target motion and the second temporary corrected instantaneous target motion that are the basis of the corrected instantaneous target motion are V and the deviation is the dynamic accuracy with respect to the moment component of the instantaneous target floor reaction force.
- the gait composed of the combination of the instantaneous target movement after correction and the instantaneous target floor reaction force can ensure the stability of the overall posture of the robot in a favorable manner.
- the first and second provisional corrected instantaneous target motions relate to the arrangement of the model elements without using the temporal change of the arrangement of the elements (first and second derivatives of the position and orientation). It can be determined by a geometric operation.
- the position and orientation of a predetermined part in the corrected instantaneous target motion can be determined by simple multiplication and addition operations.
- the instantaneous desired gait can be calculated with the floor reaction force. Correction can be made appropriately so as to achieve both the improvement of the dynamic accuracy and the fluctuation of the posture of a predetermined part such as the upper body of the robot as much as possible. A gait that enables the gait can be generated.
- the first, second, and third arrangements are actually obtained as long as the corrected instantaneous target movement as described above is consequently determined. It is not always necessary to actually calculate the translational force component and the moment component of the resultant force of the inertial forces of the above-described elements. Not necessary.
- the second invention of the mobile robot gait generator of the present invention includes an instantaneous gait generating means for sequentially generating an instantaneous target gait comprising the instantaneous target motion of the mobile robot and the instantaneous desired floor reaction force.
- a tentative gait generator a temporary correction motion determining means for determining a temporary correction instantaneous target motion obtained by temporarily correcting the position and orientation of a predetermined part of the mobile robot from the instantaneous target motion
- a target movement correcting means for determining a corrected instantaneous target movement by performing a true correction of the position and the posture of the predetermined part.
- the whole or a part of the mobile robot is represented by a model including a plurality of elements, using at least one of a rigid body having inertia and a mass point as an element.
- a model including a plurality of elements, using at least one of a rigid body having inertia and a mass point as an element.
- the first arrangement is determined from the provisional corrected instantaneous target movement determined by the provisional corrected movement determination means in accordance with a predetermined second geometric constraint condition that defines the instantaneous movement of the mobile robot and the arrangement of each element of the model.
- the arrangement of each element of the model to be determined is a second arrangement, and from the corrected instantaneous target movement determined by the target movement correcting means, the model is determined according to the second geometric constraint condition.
- the provisional corrected motion determining means may regard the difference between the arrangement of each element of the model between the second arrangement and the first arrangement as acceleration.
- the temporary correction instantaneous target is set so that the translational force component of the resultant force of the inertial force of each element calculated by the above becomes substantially zero and the moment component generated around the predetermined point becomes substantially a predetermined value.
- the target motion correcting means determines a motion, and multiplies a posture of the predetermined portion in the temporary corrected instantaneous target motion by a predetermined weight wl, and calculates the motion in the instantaneous target motion generated by the instantaneous gait generating means.
- the sum of the posture of the predetermined part multiplied by the predetermined weight w2 is determined as the instantaneous target posture of the predetermined part in the corrected instantaneous target motion, and the third arrangement and the first arrangement are determined.
- the first geometric constraint condition and the second geometric constraint condition are appropriately set, and a model is configured.
- the difference between the second arrangement and the first arrangement is determined by the provisional corrected instantaneous target movement (the respective parts of the robot determined by the provisional corrected instantaneous target movement).
- the instantaneous desired floor reaction force generated by the instantaneous gait generating means (the translational force and moment of the floor reaction force acting on the robot).
- At least one of the instantaneous target values) can be made to correspond to the degree (degree) of the dynamic error.
- the difference between the third arrangement and the first arrangement is determined by the corrected instantaneous target movement (at least the position and posture of each part of the robot determined by the target movement correcting means! It is possible to correspond to the degree (degree) of the dynamic error between the instantaneous target value of the deviation and the instantaneous target floor reaction force generated by the instantaneous gait generating means. Supplementally, in these correspondences, there is generally a stationary offset similarly to the first invention.
- the difference between the arrangement of each element of the model between the second arrangement and the first arrangement is regarded as acceleration, and the inertial force of each element calculated by regarding the difference as the acceleration.
- the position and the position of the predetermined portion are set such that the translational force component of the resultant force is substantially zero and the moment component generated around the predetermined point is substantially a predetermined value (a certain offset value).
- the provisionally corrected instantaneous target movement is determined assuming that the posture is temporarily corrected from the instantaneous target movement generated by the instantaneous gait generating means.
- This temporary corrected instantaneous target movement is equivalent to the first temporary corrected instantaneous target movement in the first aspect of the invention, and has dynamic accuracy with respect to both the translational force component and the moment component of the instantaneous target floor reaction force. Will increase. However, the posture of the predetermined portion in the temporary corrected instantaneous target movement may frequently change.
- the corrected instantaneous target movement as the true corrected instantaneous target movement of the instantaneous target movement generated by the instantaneous gait generating means is the temporary corrected instantaneous target movement based on the posture of the predetermined portion. And a posture obtained by multiplying the posture of the predetermined part by the predetermined weight wl and the posture of the predetermined part in the instantaneous target motion generated by the instantaneous gait generating means.
- the state equal to the sum of the posture multiplied by the predetermined weight w2 i.e., the arrangement of the model elements corresponding to the predetermined part, the weight wl is added to the posture of the predetermined part in the temporary corrected instantaneous target movement
- the resultant component of the inertial force of each element calculated by considering the difference between the arrangement of each element of the model and the acceleration of the respective elements in the model becomes approximately the predetermined value as the moment component generated around the predetermined point. Is determined as follows.
- the posture of the predetermined part is determined by calculating the posture obtained by multiplying the posture of the predetermined part in the provisional corrected instantaneous target movement by a weight wl and the moment generated by the instantaneous gait generating means.
- the element of the model element determined according to the second geometric condition is determined.
- the corrected instantaneous target motion has an improved dynamic accuracy with respect to the moment component of the instantaneous target floor reaction force. Good dynamic accuracy can be ensured even for the translational force component of the floor reaction force.
- the posture of the predetermined part is obtained by multiplying the posture of the predetermined part in the provisionally corrected instantaneous target movement by the weight wl and the posture of the predetermined part in the instantaneous target movement generated by the instantaneous gait generating means by the weight w2. By setting the weights wl and w2 appropriately, the fluctuation of the posture is more restricted (suppressed) than the provisionally corrected instantaneous target movement.
- the dynamic accuracy between the instantaneous target floor reaction force and the instantaneous target floor reaction force is improved better than the instantaneous target motion generated by the instantaneous gait generating means, while suppressing excessive fluctuation of the instantaneous target posture of the predetermined portion.
- the corrected instantaneous target motion that can be secured can be determined.
- the corrected instantaneous target motion and the momentary component of the instantaneous target floor reaction force have good dynamic accuracy with respect to the translational force component.
- a gait composed of a set of forces can better ensure the stability of the overall posture of the mobile robot.
- the temporary corrected instantaneous target movement and the corrected instantaneous target movement are It can be determined by the geometric operation processing on the arrangement of the element without using the temporal change of the arrangement of the element of the Dell (the first and second derivatives of the position and orientation).
- the instantaneous desired gait can be compared with the floor reaction force. Correction can be made appropriately so as to achieve both improved dynamic accuracy and minimal fluctuations in the posture of a predetermined part such as the upper body of the mobile robot, and stable movement of the mobile robot. A gait that can be performed can be generated.
- the first, second, and third arrangements are actually obtained as long as the corrected instantaneous target motion as described above is consequently determined.
- the predetermined weights wl, w2 may be basically set to weights of magnitudes (for example, 0.3, 0.7, etc.) in the range of 0 to 1, If the magnitude is constant, the horizontal component force of the translational inertial force of the overall center of gravity of the robot in the corrected instantaneous target motion, depending on the motion form and the road surface condition of the mobile robot, the friction between the mobile robot and the floor There is a possibility that the floor reaction force may not be balanced with the horizontal component generated by the force. This is the same in the second invention.
- both the predetermined weight wl and the predetermined weight w2 have a size in a range of 0 to 1, and Means for variably determining the constant weight wl in accordance with the road surface condition and Z or the motion mode of the mobile robot in accordance with the target gait according to the target gait. It is preferable to provide (third invention). In this case, it is preferable that the sum of the magnitude of the predetermined weight wl and the magnitude of the predetermined weight w2 is 1 (fourth invention).
- the weights wl and w2 are variably set as described above, so that the posture of a predetermined part of the mobile robot can be adjusted to a road surface state or a posture suitable for the movement form of the mobile robot. In this way, the instantaneous target motion after the correction that can secure the dynamic accuracy well can be determined. Further, in the fourth invention, the sum of the magnitude of the weight wl and the magnitude of the weight w2 is set to 1. Thus, the position of the predetermined part in the corrected instantaneous target movement is the position of the predetermined part in the first temporary corrected instantaneous target movement and the position of the predetermined part in the second temporary corrected instantaneous target movement.
- the dynamic accuracy between the instantaneous target motion after correction and the moment component of the instantaneous target floor reaction force and the translational force component between the corrected instantaneous target motion and the instantaneous target floor reaction force are calculated.
- the instantaneous target motion after the correction can be determined while maintaining a good balance with the dynamic accuracy of the motion.
- the posture of a predetermined part of the mobile robot is restricted to a posture suitable for the road surface condition and the movement form of the mobile robot, and the dynamic accuracy is also good.
- the instantaneous target motion after correction that can be secured at the same time can be determined.
- the motion mode include a motion mode such as walking and running of the mobile robot.
- the road surface state includes a friction state such as a friction coefficient of a road surface (floor surface).
- the predetermined weight wl may be a simple real value, but may be a weight having a frequency characteristic with respect to the posture of the predetermined part multiplied by the real weight.
- the fact that the weight wl refers to the frequency characteristics more often means that the gain (speaker) of the weight wl for each frequency component when the time series of the posture of a predetermined part is represented in the frequency domain changes according to the frequency. means.
- Such weight wl is generally expressed by a transfer function using a complex number, and functions as a filter.
- the posture of the predetermined portion in the first temporary corrected instantaneous target motion in the first invention, or the temporary corrected instantaneous target motion in the second invention By removing a required frequency component from the posture of the predetermined part in the above, the posture of the predetermined part in the corrected instantaneous target motion can be determined.
- the frequency characteristic of the weight wl is, for example, a low-cut characteristic (a characteristic that cuts off low-frequency components), the first provisionally corrected instantaneous target movement (the first invention and the third and third subordinates thereof) 4 invention) or the posture of the predetermined part in the temporary corrected instantaneous target movement (the second invention and the third and fourth inventions subordinate thereto), the steady offset ( If an error occurs, it can be removed.
- the frequency characteristic of the weight wl is changed to, for example, a high cut characteristic (a high frequency component is cut off).
- the first temporary corrected instantaneous target motion (the first invention and the third and fourth inventions subordinate thereto) is the temporary corrected instantaneous target motion (the second invention and the subordinate second invention).
- the third and fourth aspects of the present invention in the case where fine power and vibration are generated in the posture of the predetermined portion, it can be removed.
- a component caused by a difference in arrangement (posture difference) of an element (rigid body) having inertia of the model is a component of the element. It is equivalent to the product of the difference in attitude (difference in inclination angle) and the value of the inertia of the element.
- the components resulting from the difference in the arrangement (position difference) of the elements having the mass of the model are those when the difference in the position and the distance of the element from the predetermined point are represented by vectors, respectively. This is equivalent to the product of the vector (outer product) multiplied by the mass of the element.
- the component resulting from the difference in the arrangement of the elements having mass is a line segment connecting one of the two positions related to the difference between the positions and the predetermined point, and
- the angle corresponds to the angle formed by the line segment connecting the other of the two positions and the predetermined point (more specifically, the force monotonically increases or decreases according to the angle).
- the sixth invention of the present invention is the first-fifth invention, wherein, among the moment components related to a difference in arrangement of each element between the second arrangement and the first arrangement, Components of each element having the mass of the model due to the difference between the position A in the first arrangement and the position B in the second arrangement are a line segment connecting the predetermined point and the position A, It is calculated from an angle formed by the line segment connecting the predetermined point and the position B by using a substantially monotonic function related to the angle, and each angle between the third arrangement and the first arrangement is calculated.
- a component caused by a difference between the position A in the first arrangement and the position C in the third arrangement of each element having the mass of the model is the predetermined value.
- a line segment connecting the point A and the position A, and a line segment connecting the predetermined point and the position C is Characterized in that it is calculated using a force to angular force said monotonic function.
- the instantaneous target motion generated by the instantaneous gait generating means is a dynamic model representing a relationship between a motion of the mobile robot and a floor reaction force.
- the inertia force generated by a specific motion component of one or more specific parts of the mobile robot is determined using a dynamic model constructed as being substantially zero, and at least one of the specific forces described above is determined. It is suitable when it contains an element corresponding to one site (the seventh invention).
- the instantaneous target motion is constructed such that an inertial force generated by a specific motion component (a translational motion or a rotational motion in a certain direction) of one or more specific portions of the mobile robot is substantially zero.
- the instantaneous gait generation means generates the instantaneous desired motion when the specific gait generates a relatively large inertia force.
- the dynamic accuracy between the momentary floor reaction force and the instantaneous target floor reaction force tends to decrease.
- an element corresponding to at least one of the specific parts is included in the model! /, So that the distance between the corrected instantaneous target motion and the instantaneous target floor reaction force is increased. Kinetic accuracy can be accurately increased.
- the instantaneous target motion generated by the instantaneous gait generating means is based on a predetermined dynamic model representing a relationship between the motion of the mobile robot and a floor reaction force. It is determined to satisfy the desired floor reaction force or the desired ZMP, and is generated from the instantaneous desired motion by a temporal change in the arrangement of each element of the model determined according to the first geometric constraint. From the floor reaction force that balances the resultant of the inertial forces of the respective elements and the instantaneous target movement, which are generated by the temporal change in the arrangement of the respective elements of the model determined according to the second geometric constraint.
- Ru restrictive condition is set (eighth invention).
- the dynamic error between the instantaneous desired motion generated by the instantaneous gait generating means and the instantaneous desired floor reaction force is calculated from the instantaneous desired motion by the first geometrical motion. Difference between the arrangement of each element of the model determined according to the geometric constraint condition and the arrangement of each element of the model determined according to the second geometric constraint condition from the instantaneous target motion (the mass point of the model). (Or a difference in the posture of the rigid body of the model).
- the first temporary correction instantaneous target motion and the second temporary By determining the correct instantaneous target movement and further determining the corrected instantaneous target movement from these temporary corrected instantaneous target movements, the dynamics between the momentary gait generating means and the instantaneous desired floor reaction force are generated. It is possible to appropriately determine the corrected instantaneous target motion that can increase the target accuracy more than the instantaneous target motion generated by the instantaneous gait generating means, and to suppress the change in the posture of the predetermined portion. Similarly, as described above, the temporary corrected instantaneous target motion according to the second invention is determined, and the corrected instantaneous target motion having the same posture as the posture of the predetermined part in the temporary corrected instantaneous target motion is determined.
- the corrected instantaneous target motion that can increase the dynamic accuracy between the instantaneous gait generating means and the instantaneous desired floor reaction force generated by the instantaneous gait generating means can be appropriately adjusted. In addition to the determination, it is possible to suppress a change in the posture of the predetermined portion.
- the instantaneous target motion generated by the instantaneous gait generating means is defined by a predetermined dynamic model representing a relationship between the motion of the mobile robot and a floor reaction force.
- the difference between the instantaneous target motion and the total center of gravity of the arrangement of each element of the model determined according to the second geometric constraint condition multiplied by the total mass of the element is the power in the instantaneous target motion.
- the first and second geometric constraints may be set so as to substantially match the error of the overall center of gravity of the dynamic model multiplied by the total mass of the dynamic model. invention).
- the dynamic accuracy between the first temporarily corrected instantaneous target motion according to the first invention and the instantaneous target floor reaction force, and the second temporary corrected instantaneous target motion and the instantaneous target motion are calculated.
- the effect of the error of the overall center of gravity of the dynamic model which is one of the factors that lower the dynamic accuracy between the floor reaction force and the dynamic model, can be canceled.
- the dynamic accuracy between the temporarily corrected instantaneous target motion and the instantaneous target floor reaction force according to the second invention, and the dynamic accuracy between the corrected instantaneous target motion and the instantaneous target floor reaction force can also be canceled for the target accuracy.
- the mobile robot when the mobile robot is a robot including a plurality of legs or a plurality of arms extended from an upper body as a plurality of movable bodies,
- the geometric constraint condition is that the model is drawn on a straight line parallel to a line segment connecting a predetermined point near the tip of each movable body and a predetermined point near the connecting portion of the movable body with the upper body. If any of the elements of the above exists, it is preferable to include a condition U (the tenth invention).
- the first geometric constraint condition includes a condition that the upper body and each movable body on the model are held in a predetermined constant posture state, 11th invention).
- the predetermined constant posture is a posture in which the upper body of the mobile robot and the plurality of movable bodies are directed substantially in the direction of a bell (twelfth invention).
- the second geometric constraint condition is determined based on an arbitrary instantaneous target motion of the moving robot and an arrangement of each element of the model determined according to the condition. Is preferably set so as to substantially coincide with the arrangement of a part corresponding to the element in the robot following the instantaneous target movement (the thirteenth invention).
- the arrangement of each element of the model determined according to the first geometric constraints from the instantaneous target operation and The difference between the arrangement of each element of the model determined according to the second geometric constraint condition from the instantaneous target motion is determined by the instantaneous target generated by the instantaneous target motion and the instantaneous gait generating means. It is possible to appropriately cope with the dynamic error between the floor reaction force
- the mobile robot includes a plurality of legs or a plurality of arms extending from the upper body as a plurality of movable bodies, and the upper body of each movable body. And a middle part between the connecting part of the movable body and the tip of the movable body, and a momentary target movement generated by the momentary gait generating means is a movement of the robot and a floor reaction force. And a dynamic model constructed on the assumption that the inertial force generated at or near the intermediate portion of each movable body due to the bending and stretching motion of each movable body is almost zero. When using the model, at least the intermediate portion of each of the movable bodies Or, it is preferable that the model includes a mass point corresponding to a nearby part as an element (the fourteenth invention).
- the instantaneous target motion has an inertia force generated at or near the center of each movable body due to the bending and stretching motion of each movable body being substantially zero (ie, ignoring the inertial force).
- the instantaneous gait generating means is used to generate a desired gait such that the bending motion of each movable body is performed relatively quickly.
- the dynamic accuracy between the instantaneous target motion and the instantaneous target floor reaction force at which the moment occurs is likely to decrease.
- the first and second temporary correction instantaneous moments according to the first invention are included by including, as an element, a mass point corresponding to the intermediate portion of each movable body or a portion in the vicinity thereof.
- the joint at the intermediate portion of each movable body is determined.
- the instantaneous target movements can be determined by compensating for the influence of the inertial force accompanying the bending and stretching movements of the movable body due to the bending movement of the movable body.
- the dynamic accuracy between the determined instantaneous target motion and the instantaneous target floor reaction force determined by the instantaneous target gait generating means can be increased.
- the instantaneous gait composed of the instantaneous corrected desired movement according to the first or second invention and the instantaneous desired floor reaction force is an instantaneous gait while suppressing a change in the posture of the predetermined portion. Dynamic accuracy can be higher than the instantaneous gait generated by the generating means.
- the first geometric constraint may be set, for example, in the same manner as in the tenth or eleventh invention, and the second geometric constraint may be set in the thirteenth invention. Just set it as invented. In particular, it is preferable to set the first and second geometric constraint conditions as in the tenth invention and the thirteenth invention, respectively.
- the first geometric constraint condition is defined as a line segment connecting a predetermined point near the tip of each movable body and a predetermined point near the joint of the movable body with the upper body. And a condition that a mass corresponding to an intermediate portion of the movable body or a portion in the vicinity of the model exists, and the second geometric constraint condition is an arbitrary condition of the mobile robot.
- the arrangement of each element of the model determined according to the conditions from the It is preferable that the setting is made so as to substantially coincide with the arrangement of the part corresponding to the element in the robot following the time target movement (the fifteenth invention).
- each movable position in the arrangement is determined.
- a mass point of the model corresponding to the middle part of the body or a part in the vicinity thereof hereinafter, referred to as a movable body intermediate mass point
- a movable body intermediate mass point in the first arrangement which is on the line segment
- the inertial force corresponds to the error of the instantaneous desired floor reaction force generated by the instantaneous gait generating means.
- the first and second temporary corrected instantaneous target movements according to the first invention or the temporary corrected instantaneous target movement and the corrected instantaneous target movement according to the second invention, will be described with respect to the first or second invention.
- the influence of the inertial force accompanying the bending and stretching movement of the movable body due to the bending movement of the joint at the intermediate portion of each movable body is compensated, and the instantaneous target movement and the instantaneous target floor reaction force are compensated.
- the dynamic accuracy between them can be increased.
- a bipedal mobile robot is taken as an example of the mobile robot.
- FIG. 1 is a schematic diagram showing an outline of an overall configuration of a bipedal mobile robot to which an embodiment of the present invention is applied.
- a bipedal mobile robot (hereinafter, referred to as a robot) 1 includes a pair of left and right legs 2, 2 extending downward from an upper body (base of the robot 1) 3.
- the upper body 3 corresponds to a “predetermined part” in the present invention.
- the legs 2 and 2 have the same structure, each having six joints.
- the six joints, in order of the upper body 3 side force, are joints 10R and 10L for the rotation (rotation) of the crotch (lumbar) (rotation in the Y direction relative to the upper body 3) and the roll direction of the crotch (lumbar).
- Joints 12R and 12L for rotation (around the X axis), Joints 14R and 14L for rotation in the crotch (lumbar) pitch direction (around the Y axis), and joints 16R and 16L for rotation in the knee pitch direction And rotation of the ankle pitch direction And the joints 20R and 20L for rotation of the ankle in the roll direction.
- symbols R and L are symbols corresponding to the right side and the left side of the robot 1, respectively.
- the hip joint (or the hip joint) is connected to the joints 10R (L) and 12R.
- the knee joint is composed of the joint 16R (L)
- the ankle joint is composed of the joints 18R (L) and 20R (L).
- the hip joint and the knee joint are connected by a thigh link 24R (L)
- the knee joint and the ankle joint are connected by a lower leg link 26R (L).
- the "link” of the robot 1 is used to mean a part that can be regarded as a rigid body of the robot 1.
- upper body 3 is also one link (rigid body), and in that sense upper body 3 is sometimes called upper body link.
- a pair of left and right arms 5 and 5 are attached to both upper sides of the upper body 3, and a head 4 is disposed at the upper end of the upper body 3.
- Each arm 5 has a shoulder joint composed of three joints 30R (L), 32R (L) and 34R (L), an elbow joint composed of joint 36R (L), and a joint 38R (L). And a wrist part 40R (L) connected to the wrist joint. Links are formed between the shoulder joint and the elbow joint, and between the elbow joint and the wrist joint.
- the joint shown below
- the desired motion of both feet 22R and 22L can be performed.
- the robot 1 can arbitrarily move in the three-dimensional space.
- each arm 5 can perform a motion such as arm swing by rotating its shoulder joint, elbow joint, and wrist joint. [0062] As shown in Fig.
- a known 6-axis force sensor 50 is interposed between the ankle joints 18R (L) and 20R (L) of each leg 2 and the foot 22R (L). ing.
- the 6-axis force sensor 50 is for detecting whether the foot 22R (L) of each leg 2 lands, and detecting a floor reaction force (ground load) acting on each leg 2 and the like.
- the detection signals of the three-directional components Fx, Fy, Fz of the translational force of the floor reaction force and the three-directional components Mx, My, Mz of the moment are output to the control unit 60.
- the body 3 is provided with a posture sensor 54 for detecting the inclination angle and the angular velocity of the body 3 with respect to the Z axis (vertical direction (gravity direction)), and the detection signal is controlled by the posture sensor 54. Output to control unit 60.
- the attitude sensor 54 includes an acceleration sensor and a gyro sensor (not shown), and detection signals from these sensors are used to detect the inclination angle of the body 3 and its angular velocity.
- each joint of the robot 1 has an electric motor 64 (see FIG. 3) for driving the joint, and a rotation amount of the electric motor 64 (rotation angle of each joint).
- (Rotary encoder) 65 (see FIG. 3) is provided for detecting the signal, and a detection signal of the encoder 65 is output from the encoder 65 to the control unit 60.
- a joystick (operator) 73 for manipulating the robot 1 is provided outside the robot 1 which is not shown in FIG. 1.
- the joystick 73 is operated by operating the joystick 73.
- the mode of operation that specifies the direction of movement of the robot 1, such as turning the robot 1 moving straight ahead, specifying the moving direction of the robot 1, walking and running the robot 1, and the frictional state of the floor surface (road surface state). It is configured so that a request or restriction on the gait of the robot 1 such as designation can be input to the control unit 60 as needed.
- the joystick 73 can communicate with the control unit 60 by wire or wirelessly.
- FIG. 2 is a view schematically showing a basic configuration of a tip portion (including each foot 22R (L)) of each leg 2 in the present embodiment.
- a spring mechanism 70 is provided between the foot 22R (L) and the six-axis force sensor 50, and the sole (the bottom surface of each foot 22R, L) is provided.
- a sole elastic body 71 which is also strong such as rubber is attached.
- a compliance mechanism 72 is configured by the spring mechanism 70 and the sole elastic body 71.
- the spring mechanism 70 includes a rectangular guide member (not shown) attached to the upper surface of the foot 22R (L) and an ankle joint 18R (L) (in FIG. 2, the ankle joint). 20R (L) is omitted)
- a piston-like member (not shown) which is attached to the 6-axis force sensor 50 side and is finely movable and accommodated in the guide member via an elastic material (rubber or spring).
- the foot 22R (L) indicated by a solid line in FIG. 2 indicates a state when no floor reaction force is applied.
- the spring mechanism 70 of the compliance mechanism 72 and the sole elastic body 71 bend, and the foot 22R (L) moves to the position and posture as exemplified by the dotted line in the figure.
- the structure of the compliance mechanism 72 is important not only to reduce the landing impact but also to enhance controllability. The details thereof are described in detail in, for example, Japanese Patent Application Laid-Open No. 5-305584 previously proposed by the present applicant, and further description in this specification will be omitted.
- FIG. 3 is a block diagram showing a configuration of the control unit 60.
- the control unit 60 is composed of a microcomputer, and has a first arithmetic unit 90 and a second arithmetic unit 92 which also have CPU power, an AZD variable ⁇ 80, a counter 86, a DZA variable ⁇ 96, a RAM84, a ROM94, And a bus line 82 for exchanging data between them.
- the output signals of the 6-axis force sensor 50, posture sensor 54 (acceleration sensor and rate gyro mouth sensor), joystick 73, etc. of each leg 2 are converted into digital values by the AZD translator 80, and then converted to digital signals.
- the outputs of the encoders 65 (rotary encoders) of the joints of the robot 1 are input to the RAM 84 via the counter 86.
- the first arithmetic unit 90 generates a target gait as described later and calculates a joint angle displacement command (a displacement angle of each joint or a command value of a rotation angle of each electric motor 64). Send to 4.
- the second arithmetic unit 92 reads the joint angle displacement command and the actually measured joint angle detected based on the output signal of the encoder 65 from the RAM 84, and determines the amount of operation required to drive each joint. calculate. Then, the calculated operation amount is output to the electric motor 64 for driving each joint via the DZA converter 96 and the servo amplifier 64a.
- FIG. 4 is a block diagram showing a main functional configuration of the control unit 60 of the robot 1 according to the embodiment of the present specification.
- the parts other than the “real robot” part in FIG. 4 are constituted by the processing functions executed by the control unit 60 (mainly the functions of the first arithmetic unit 90 and the second arithmetic unit 92). It is.
- the processing function is performed by a program implemented in the control unit 60. This is realized by a program or the like.
- the symbols R and L are omitted when it is not necessary to particularly distinguish the left and right of each part (the leg 2, the arm 5, etc.) of the robot 1.
- the control unit 60 includes a gait generator 100 that generates and outputs a desired gait freely and in real time as described later.
- the gait generator 100 realizes the embodiment of the present invention by its function.
- the desired gait output by the gait generator 100 includes a corrected desired body posture trajectory (trajectory of the desired posture of the body 3), a corrected target body position trajectory (a trajectory of the desired position of the body 3), Foot position / posture trajectory (trajectory of target position and target posture of each foot 22), target arm posture trajectory (trajectory of target posture of each arm), target ZMP (target total floor reaction force center point) trajectory, target It consists of the corrected desired floor reaction force moment trajectory around the ZMP and the desired total floor reaction force trajectory.
- a portion (head or the like) that is movable with respect to the upper body 3 is provided in addition to the leg 2 and the arm 5, the target position / posture trajectory of the movable portion is added to the target gait.
- the “trajectory” in a gait means a temporal change pattern (time-series pattern), and is sometimes called a “pattern” instead of a “trajectory”.
- “posture” means a spatial orientation.
- the body posture is the inclination angle (posture angle) of the body 3 in the roll direction (around the X axis) with respect to the Z axis (vertical axis) and the inclination angle (posture angle) of the body 3 in the pitch direction (around the Y axis).
- the foot posture is represented by a spatial azimuth of two axes fixedly set for each foot 22.
- the body posture may be referred to as the body posture angle.
- the target arm posture regarding the arm body 5 is represented by a relative posture with respect to the body 3 in the embodiment of the present specification.
- the body position means a position of a representative point of the body 3 which is determined in a rough manner (a fixed point in a local coordinate system arbitrarily fixedly set with respect to the body 3).
- the foot position means the position of a representative point of each foot 22 that has been roughly determined (fixed point in a local coordinate system arbitrarily fixedly set for each foot 22).
- the representative point of each foot 22 is on the bottom surface of each foot 22 (more specifically, the point where the perpendicular from the center of the ankle joint of each leg 2 to the bottom surface of each foot 22 intersects the bottom surface). Is set to
- the corrected target body posture and the corrected target body position for the body 3 are based on a certain basic target body posture (temporary target body position) and a target body position (temporary target body position). correction It was done.
- the basic target body position / posture corresponds to a displacement dimension corrected body position / posture described later.
- the components of the gait other than the components relating to the floor reaction force that is, the components relating to the position and orientation of each part of the robot 1, such as the foot position and orientation, the body position and orientation, are collectively referred to as “movement”. Movement ".
- the floor reaction force acting on each foot 22 (floor reaction force composed of translational force and moment) is called “each foot floor reaction force”, and the robot 1 has two (two) foot reaction forces for the foot 22R and 22L.
- the resultant force of each foot floor reaction is called the total floor reaction.
- each floor floor reaction force is hardly mentioned, and unless otherwise noted, “floor reaction force” is treated as synonymous with “total floor reaction force”.
- the desired floor reaction force is generally represented by an action point and a translational force and a moment acting on the point. Since the action point is good for everywhere, countless expressions can be considered for the same desired floor reaction force.However, in particular, the target floor reaction force center point (the target position of the center point of the total floor reaction force) is used as the working point, When expressing the reaction force, the moment component of the desired floor reaction force is zero except for the vertical component (moment component around the vertical axis (Z axis)). In other words, the horizontal component of the moment of the desired floor reaction force around the center point of the desired floor reaction force (the moment about the horizontal axis (X axis and Y axis)) becomes zero.
- the target motion orbit force of the robot 1 is calculated as ZMP (the target motion orbit force is also calculated and the resultant force of inertia and gravity acts around that point).
- ZMP the target motion orbit force is also calculated and the resultant force of inertia and gravity acts around that point.
- the moment when the moment becomes zero except for the vertical component) and the desired floor reaction force center point coincide with each other, so it is the same as giving the desired ZMP trajectory instead of the desired floor reaction force center point trajectory.
- the vertical position When the body height is determined, the translational floor reaction force vertical component is determined dependently. Furthermore, the horizontal position trajectory of the robot 1 (or the position trajectory of the overall center of gravity) is adjusted so that the resultant force of the inertia and gravity due to the motion of the desired gait becomes zero in the horizontal component of the moment generated around the target ZMP. By determining the road, the translational floor reaction force horizontal component is also determined dependently. Therefore, when the robot 1 walks, only the target ZMP may be the only physical quantity that should be explicitly set for the floor reaction force of the target gait.
- the translational floor reaction force vertical component also This is important for operation control. For this reason, it is desirable to determine the trajectory of the robot 1 such as the target body vertical position after explicitly setting the target trajectory of the translational floor reaction force vertical component. Also, when walking the robot 1, if the coefficient of friction is low and the robot 1 moves on the floor (low mu road), the translational floor reaction force vertical component (more strictly, the translational floor reaction force Since the component perpendicular to the floor affects the frictional force, it is desirable to explicitly set the target trajectory of the translational floor reaction vertical component to prevent the robot 1 from slipping. Further, in the embodiment of the present invention, in the desired gait finally output by the gait generator 100, a corrected desired floor reaction force moment (a moment whose horizontal component is not always 0) is generated around the desired ZMP.
- the corrected target floor around the target ZMP is used as a component relating to the floor reaction force of the target gait output by the gait generator 100.
- the reaction moment and the target translational floor reaction force vertical component are included!
- the desired gait output by the gait generator 100 is, in a broad sense, “a set of a desired motion trajectory and a desired floor reaction force trajectory during one or more steps”. In a narrow sense, it is used as a meaning of ⁇ a set of a target motion trajectory during one step and a target floor reaction force trajectory including a target ZMP, a corrected target floor reaction force moment, and a target translational floor reaction force vertical component. '' Used in a meaning.
- a desired gait provisional desired gait created in the process of determining a final desired gait (a desired gait output by the gait generator 100).
- the horizontal component of the desired floor reaction force moment around the target ZMP is set to 0 as defined by the original target ZMP. Therefore, in temporary provisional gaits other than the finally determined target gait (simplified model gait, first provisional corrected gait, second provisional corrected gait, and displacement dimension corrected gait described later), The corrected desired floor reaction force moment is subtracted from the desired gait of Used in taste.
- a desired gait (temporary desired gait) created in the process of determining a final desired gait (the desired gait output by the gait generator 100) is a real gait. It is closely related to the invention. For this reason, most of the desired gaits in the following description are used in the meaning of the strictly defined desired gait minus the corrected desired floor reaction force moment (gaits satisfying the desired ZMP). You.
- the "floor reaction force vertical component” means “translational floor reaction force vertical component”
- the vertical component of the moment of the floor reaction force (the component around the vertical axis) is The term “moment” is used to distinguish it from the “floor reaction force vertical component”.
- the “floor floor reaction force horizontal component” means “translational floor reaction force horizontal component”.
- one step of the desired gait is used in the meaning from the time when one leg 2 of the robot 1 lands to the time when the other leg 2 lands.
- the two-leg supporting period in the gait is a period in which the robot 1 supports its own weight with the two legs 2, 2, and the one-leg supporting period is the robot 1's own weight with only one of the legs 2.
- the aerial period refers to the period in which both legs 2 and 2 are separated from the floor (floating in the air).
- the leg 2 that does not support the weight of the robot 1 during the one-leg supporting period is called a free leg. Note that there is no double leg support period in the running gait of the robot 1 in which the one leg support period and the aerial period are alternately repeated.
- both the legs 2 and 2 do not support the weight of the robot 1 during the aerial period, but for convenience, the leg 2 and the supporting leg, which were the free legs during the one-leg supporting period immediately before the aerial period, are not used.
- the legs 2 are also referred to as a free leg and a supporting leg during the aerial period, respectively.
- the trajectory of the desired gait is described in a global coordinate system (a coordinate system fixed to the floor).
- a global coordinate system for example, a support leg coordinate system determined corresponding to the landing position / posture of the support leg foot 22 is used.
- a vertical line extending from the center of the ankle joint to which the foot 22 is connected to the floor surface is connected to the floor, for example, with almost the entire bottom surface of the support leg foot 22 grounded to the floor.
- a coordinate system in which the point of intersection is the origin and the front and rear directions of the foot 22 when projected on a horizontal plane passing through the origin is the X axis direction and the left and right direction is the Y axis direction (the Z axis direction is vertical Direction).
- FIG. 5 is a block diagram showing details of gait generator 100. With reference to FIG. 5, a more specific outline of the processing of the gait generator 100 will be described below. [0087] As shown, the gait generator 100 includes a gait parameter determination unit 100a. The gait parameter determination unit 100a determines a gait parameter value or a time-series table that defines a desired gait.
- the gait parameters determined by the gait parameter determining unit 100a include, among the desired gaits, a desired foot position / posture trajectory, a desired arm posture trajectory, a desired ZMP trajectory, And a parameter defining the desired floor reaction force vertical component trajectory, respectively.
- the gait generator 100 when the gait generator 100 generates the desired gait, the basic gait generation for the gait such as the expected landing position / posture and landing time of the free leg foot 22, the expected landing time, or the stride, moving speed, etc.
- the required values are given to the joystick 73 or a device gait generator 100 such as an unillustrated action plan unit (a device for creating an action plan for the robot 1).
- the gait generator 100 also reads the required parameters from a storage medium that previously stores the required parameters. Then, the gait parameter determination unit 100a of the gait generator 100 determines a gait parameter according to the required parameter.
- the gait parameters determined by the gait parameter determining unit 100a include a reference body posture trajectory, a ZMP allowable range, and a floor reaction force horizontal component allowable range, respectively. It also includes the parameters to be specified.
- the reference body posture trajectory is not finally output by the gait generator 100, but is taken into consideration when determining a desired gait.
- the reference body posture trajectory is given from the joystick 73 or the action planning unit with respect to the body posture of the robot 1 or directly follows a predetermined request (a request for maintaining the body posture in a vertical posture). This is the generated body posture trajectory.
- the desired body posture hereafter, ⁇ body posture '' without the ⁇ reference '' indicates the desired body posture
- the desired gait generates a corrected desired floor reaction force moment (which is generally not 0) around the desired ZMP. Will be corrected as follows. Therefore, the target ZMP is different from the original definition (the definition that the horizontal component of the floor reaction force moment is zero), and ZMP (hereinafter, true ZMP) that satisfies the original definition is , The corrected desired floor reaction force moment is calculated using the desired floor reaction force vertical component. Move to a position shifted from the target ZMP by the divided value.
- the true ZMP of the corrected gait (the desired gait finally output by the gait generator 100) is at least the ZMP possible range (so-called supporting polygon. Floor and bottom of foot 22) And the floor reaction force action point (ZMP), assuming that no adhesive force acts between and. Furthermore, in order to secure a sufficient stability margin of the robot 1, it is desirable that the true ZMP of the corrected gait is in the vicinity of the center of the ZMP possible range. Therefore, in the embodiment of the present specification, an allowable range in which a true ZMP of the corrected gait can exist is set. This range is called the ZMP allowable range.
- the ZMP allowable range is set to match the ZMP possible range or to be included in the ZMP possible range.
- the value obtained by dividing the corrected desired floor reaction force moment around the target ZMP by the vertical component of the desired floor reaction force represents the amount of deviation of the true ZMP position with respect to the target ZMP.
- the deviation amount of the true ZMP position from the target ZMP ZMP converted value of the corrected target floor reaction force moment
- the ZMP allowable range can be converted into the allowable range of the corrected target floor reaction force moment using the boundary position and the target floor reaction force vertical component, and the allowable range of the corrected target floor reaction force moment can be converted. May be set instead of the ZMP tolerance.
- the floor reaction force horizontal component allowable range is a floor reaction force horizontal component that can generate a frictional force on the contact surface of the foot 22 of the robot 1 with the floor so that the foot 22 does not slip. It is the allowable range of the component.
- at least the motion of the desired gait (the desired motion) finally output by the gait generator 100 is the floor reaction force generated by the motion, which is in proportion to the water equivalent of the inertial force of the robot 1.
- the horizontal component is generated so that it falls within the floor reaction force horizontal component allowable range.
- the gait parameters determined by the gait parameter determination unit 100a are input to the target instantaneous value generation unit 100b. Based on the input gait parameters, the target instantaneous value generation unit 100b instantaneously activates some components of the target gait, such as a reference body posture, a desired foot position / posture, a desired ZMP, and a desired floor reaction force vertical component. Value (value for each predetermined control processing cycle of the control unit 60) It is calculated (generated) sequentially. In FIG. 5, only some of the target instantaneous values are representatively shown.
- the target instantaneous value calculated by the target instantaneous value generator 100b is input to the simplified model gait generator 100c.
- the simple dani model gait generation unit 100c based on the input target instantaneous values, dynamically expresses the relationship between the motion of the robot 1 and the floor reaction force as described below (hereinafter, a simplified model and! /
- the instantaneous value of the target body position / posture (temporary target body position / posture) is calculated using
- the simple-model model gait generator 100c is configured to satisfy the dynamic equilibrium condition on the simplified model, that is, the inertial force generated by the target motion of the robot 1 on the simple model.
- the instantaneous value of the desired body position and posture depends on the inertial force generated by the desired motion and the effect on the robot 1.
- the resultant force with the gravitational force becomes the moment horizontal component force generated around the target ZMP ⁇ , and the translational force vertical component of the resultant force (in other words, the inertial force due to the vertical translational motion of the entire center of gravity of the robot 1 (The resultant force with gravity) is determined so as to balance the vertical component of the desired floor reaction force.
- the instantaneous value of the desired gait (temporary desired gait) including the desired body position / posture is sequentially determined.
- a desired gait that has the desired body position / posture obtained by the simple dani model gait generator 100c as a component is referred to as a simplified model gait.
- the target instantaneous values input to the simplified model gait generator 100c need not be all the target instantaneous values calculated by the target instantaneous value generator 100b.
- the input required for the simple model model gait generator 100c depends on the structure of the simplified model or the constraint conditions appropriately added thereto. For example, in FIG. 5, the target foot position / posture is input to the simple model model gait generator 100c. However, in the simplified model in the first embodiment described later, it is necessary to input the target foot position / posture. There is no.
- the simple dani model gait generator 100c together with the target instantaneous value generator 100b, constitutes an instantaneous gait generator in the present invention.
- the target body position / posture calculated by the simple model model gait generator 100c is a displacement dimension step. Input to the capacity correction unit lOOd.
- the displacement dimension gait corrector lOOd is also input with the target body position / posture force, the instantaneous value of the desired foot position / posture, and the instantaneous value of the target ZMP.
- it is not essential to input the target ZMP to the displacement dimension gait corrector lOOd but a center point relating to the angular momentum product described later is generally input.
- the target ZMP is input to the displacement dimension gait corrector lOOd as an example of the center point.
- the displacement dimension gait corrector lOOd uses the first and second displacement dimension correction models described later based on the input instantaneous values of the desired body position and orientation, and simplifies the model gait.
- the instantaneous value of the displacement dimension corrected body position / posture obtained by correcting the target body position / posture obtained by the generator 100c is obtained.
- the first and second displacement dimension correcting models are generally a model configured by using at least one of a mass point and a link having inertia as elements (geometric model). ), And the arrangement (position of mass point, link posture) force of each element can correspond to the position and posture of one or more parts in the instantaneous motion of the robot 1.
- both the first and second displacement dimension correcting models are composed of the same element.
- different geometrical constraint conditions are defined for the arrangement of the elements, and any instantaneous target movement of the robot 1 (each part of the robot 1).
- the arrangement of the elements of each displacement dimension correcting model corresponding to the instantaneous value is determined based on each different geometric constraint condition. Therefore, when a given instantaneous motion is given, the arrangement of the elements of each displacement dimension correcting model corresponding to the given instantaneous motion is generally different from each other.
- the displacement dimension gait correction unit lOOd calculates the difference in the arrangement of the elements (difference in the position of the mass point or difference in the attitude angle of the link) in the first and second displacement dimension correction models.
- the target body position and posture of the simplified model gait are corrected, and the instantaneous values of the displacement dimension corrected body position and posture are sequentially obtained.
- the displacement dimension gait correction unit lOOd is requested from the joystick 73 or the action planning unit (not shown) of the robot 1 in the target gait.
- the movement mode of the robot 1 eg, walking and running of the robot 1
- the operation mode of the robot 1 indicating the friction state of the floor surface eg, the magnitude of the friction coefficient
- the displacement dimension gait correction unit lOOd performs the displacement dimension correction body position appearance according to the input operation mode.
- the instantaneous value of the force is variably obtained.
- the operation mode includes a traveling mode in which the robot 1 travels and a low-friction floor surface in which the robot 1 walks on a floor surface with a relatively small friction coefficient (on a low mu road).
- a walking mode and a normal mode which is an operation mode other than these modes.
- the normal mode includes an operation mode in which the robot 1 walks on a floor having a relatively large coefficient of friction (normal floor).
- the displacement dimension gait corrector lOOd configures first temporary corrected motion determining means, second temporary corrected motion determining means, and target motion correcting means in the first invention, depending on its function, or It constitutes provisional correction motion determining means and target motion correction means in the present invention.
- the instantaneous value of the displacement dimension corrected body position / posture obtained by the displacement dimension gait correction unit lOOd is input to the full model correction unit 100e.
- the target instantaneous values calculated by the target instantaneous value generation unit 100b (excluding the instantaneous values of the reference body position and orientation) are input to the full model correction unit 100e.
- the full model correction unit 100b calculates a corrected target body position and orientation obtained by correcting the displacement dimension body position and orientation using a full model as a dynamic model with higher dynamic accuracy than the simplified model.
- a corrected target floor reaction chamois which is a target value of the floor reaction force moment horizontal component around the target ZMP, is calculated.
- the full model correction unit 100e executes the processing of E1 or E2 so as to satisfy the following conditions of D1 to D3. That is, the full model correction unit 100e
- a gait generated using the simple model is corrected using a displacement dimension correcting model (hereinafter referred to as a displacement dimension corrected gait). Satisfies dynamic equilibrium conditions with accuracy.
- the true ZMP (ZMP that satisfies the original definition modified by generating a corrected desired floor reaction force moment around the target ZMP) is the ZMP allowable range (the allowable range where sufficient stability margin can be maintained).
- the displacement dimension correction gait corrects the body position and orientation, and outputs a corrected desired floor reaction force moment around the target ZMP (corrects the desired floor reaction force).
- the process of E2 is executed so as to satisfy the conditions of D1 to D3.
- the processing of the full model correction unit lOOe in the embodiment of the present specification is described in detail in, for example, PCT International Publication WOZ03Z057427ZA1 previously proposed by the present applicant (specifically, This is the same as in step S038 in FIG. 13). Therefore, a detailed description of the processing of the full model correction unit 100e in this specification will be omitted.
- instantaneous values of the desired gait including the corrected desired body position and orientation determined as described above, the corrected desired floor reaction force moment around the target ZMP, and the instantaneous values of the desired foot position and orientation.
- the composite compliance control device 101 controls the joint actuator (electric motor 64) so as to follow a desired gait while maintaining the balance of the robot 1. Note that more specific processing of the composite compliance control device 101 will be described later.
- the outline of the gait generator 100 has been described above. Note that the outline of the gait generator 100 described above is the same as in the other embodiments of the present specification.
- the first embodiment is an embodiment of the first, third, fourth, seventh, eleventh, and eleventh aspects of the invention.
- FIG. 6 shows the structure of the simple model in the first embodiment.
- the simplified model is a one-mass model including one mass point (upper mass point) 3 m corresponding to the upper body 3 of the robot 1.
- the robot 1 shown in FIG. 6 is a schematic view of the robot 1 viewed from the side, and the illustration of the arms 5, 5 and the head 6 is omitted.
- FIG. 6 including the drawings of the embodiments other than the first embodiment
- the robot 1 when the robot 1 is illustrated, it is the same as FIG. , Omit arms 5, 5 and head 6 To do.
- the X axis and the Z axis described in the subsequent drawings including FIG. 6 indicate a global coordinate system.
- the upper body mass point 3m of the simple dani model in Fig. 6 is a point uniquely determined in accordance with the position and orientation of the upper body 3, that is, a point in the local coordinate system arbitrarily fixedly set to the upper body 3.
- a fixed point (a point having a predetermined positional relationship with the representative point of the upper body 3 in the oral coordinate system) is set.
- the mass of the upper body mass 3 m is the same as the total mass mtotal of the robot 1. Note that the body mass point 3m may coincide with the representative point of the body 3, but is generally different.
- the dynamics of this simplified model is expressed by the dynamics of an inverted pendulum composed of an upper body mass 3m and a variable-length link 3b that swingably supports the upper mass 3m with the target ZMP as a fulcrum.
- a motion equation representing the relationship between the motion of the robot 1 and the floor reaction force in the simple model is expressed by the following equations 01, 02, and 03.
- the so-called sagittal plane is described here.
- the equation of motion on the lateral plane (the plane containing the left and right axes (Y axis) and the vertical axis (Z axis), V, so-called frontal plane) is omitted.
- d2X / dt2 for a variable X means the second derivative of the variable X.
- variables related to the dynamics of the simple model of FIG. 6 are defined as follows.
- g gravity acceleration
- Zb vertical position of the body mass point
- Xb horizontal position of the body mass point
- mtotal total mass of robot 1
- Fx floor reaction force horizontal component (specifically, translational floor reaction force Front-rear direction (X-axis) component)
- Fz Floor reaction force vertical component (specifically, vertical (Z-axis) component of translational floor reaction force)
- My Floor reaction force moment around target ZMP (specifically, floor reaction force) Moment about left and right axis (Y axis)
- Xzmp Horizontal position of target ZMP
- Zzmp Vertical position of target ZMP.
- the simplified model of the first embodiment described above is a one-mass model in which only the upper body 3 has a mass point 3m, so that the inertial force generated by the motion of each leg 2 and the Inertia (moment of inertia) is ignored.
- the simplified model of the first embodiment is a dynamic model constructed by assuming that the inertial force generated by the movement (translation or posture change movement) of each leg 2 and the posture change movement of the upper body 3 is 0. It can be said that there is.
- the simplified model in each of the embodiments of the present specification including the first embodiment is generated by a specific motion (translation motion, posture change motion, etc.) of at least one or more specific portions of the robot 1. It is constructed assuming that the inertial force is almost 0 (ignoring the inertial force).
- the simplified model of the first embodiment is a one-mass model, it may be, for example, a three-mass model having a mass near the foot 22 of each leg 2. Also, for example, a model in which the upper body 3 has inertia (moment of inertia) around the upper body mass point 3m! / ⁇ .
- the diagram on the right side of (c) shows the structure of the first displacement dimension correcting model in the first embodiment, and the diagram on the left side shows the overall posture of the robot 1 corresponding to the diagram on the right. (Posture state of gait model) and the simulated model.
- the robot 1 shown on the right side of Figs. 7 (a), (b) and (c) is a robot in a standing state in which both legs 2 and 2 are arranged in the left-right direction (Y-axis direction) and Figure 1 is a side view (sagittal plane). For this reason, the two legs 2 and 2 overlap in the drawing.
- the first displacement dimension correcting model of the first embodiment includes one upper body mass Al corresponding to the upper body 3 of the robot 1 and a thigh mass point corresponding to the thigh link portion near the knee joint of each leg 2.
- This is a 5-mass model consisting of A2, A3, and foot mass points A4, A5 corresponding to the tip of each leg 2 (foot 22), respectively.
- the body 3 (body link) in the first displacement dimension correction model has an inertia (moment of inertia) lb around the body mass point A1.
- the first displacement dimension correction model is composed of the mass points A1 to A5 and the body link having inertia lb as elements.
- the body link having the mass points A2 and A5 and the inertia lb of the first displacement dimension correction model is an element that the simple model of FIG. 6 does not have, and the motion ( The body 3 generates an inertial force by its posture change movement).
- the upper body mass point A1 and the foot mass points A4 and A5 of the first displacement dimension correction model correspond to the position and orientation of the corresponding part (the upper body 3 and each foot 22).
- a fixed point on the local coordinate system arbitrarily fixed to the corresponding part (a representative point of the corresponding part on the local coordinate system of the corresponding part and a predetermined positional relationship).
- the position of the body mass point A1 of the body 3 on the local coordinate system is generally different from the body mass point 3m of the simple model shown in FIG.
- the thigh mass points A2 and A3 are set to fixed points (fixed points near the knee joint) in the local coordinate system arbitrarily fixed to the thigh link 24 of each leg 2.
- the mass of the upper body mass A1 includes the mass of the upper body 3 and the masses of the arms 5, 5 and the head 4!
- a certain geometric constraint condition is set for the arrangement of the elements of the first displacement dimension correcting model.
- the posture state of the robot 1 is such that the upper body 3 is in the vertical posture and the legs 2 and 2 are in the left-right direction (Y-axis direction) of the robot 1. It is constantly constrained to the posture state (standing upright state) standing side by side at intervals (for this reason, the model of the first displacement dimension correction model on the right side in Figs. 7 (a), (b), and (c) is In the figure, the mass points A2 and A4 corresponding to one leg 2 respectively overlap with the mass points A3 and A5 corresponding to the other leg 2).
- the relative positional relationship between the upper body mass Al, each foot mass A4, A5, and the thigh mass A2, A3 is restricted to a predetermined positional relationship corresponding to the upright posture state of the robot 1.
- the posture of the body 3 which is a link (rigid body) having inertia is restricted to a vertical posture (a posture in which the posture angle with respect to the vertical axis is 0).
- the position of each mass point Al-A5 of the first displacement dimension correction model on the global coordinate system (the coordinate system fixed to the floor) is determined by the instantaneous value of the motion of the simple gait model gait. It is determined correspondingly.
- the overall center of mass of the mass points A1 to A5 is the position of the overall center of gravity of the robot 1 on the simplified model, that is, the body of the simplified model.
- the positions of mass points A1-A5 in the global coordinate system are determined so that they match the position of point 3m (the position in the global coordinate system).
- the relative positions of the mass points A1 to A5 of the first displacement dimension correction model are constant, the position of the entire center of gravity of the mass points A1 to A5 (the position in the global coordinate system) ), The position of each mass point A1—A5 in the global coordinate system is uniquely determined
- determining the arrangement of the elements of the first displacement dimension correction model means that the arrangement of the elements of the first displacement dimension correction model (in the global coordinate system).
- the geometric constraints (1) for defining the positions of the mass points A1—A5 and the posture of the body link are defined as follows, the instantaneous motion of the simple gait model gait is used to determine the geometric constraints. This is the same as determining the arrangement of the elements of the first displacement dimension correcting model according to the bundle condition (1).
- Geometric constraint condition (1) For any given instantaneous target motion, the posture state of the robot 1 corresponding to the arrangement of the elements of the first displacement dimension correction model is constantly maintained in the upright posture state. In addition, the overall center of gravity of the element of the first displacement dimension correcting model matches the overall center of gravity of the robot 1 in the given instantaneous target motion.
- the geometric constraint (1) is the first geometric constraint in the present invention. This corresponds to the constraint condition.
- FIG. 8 shows the structure of the second displacement dimension correcting model.
- the second displacement dimension correction model has the same components as the first displacement dimension correction model, and has five mass points A1 to A5 as in the first displacement dimension correction model.
- Body 3 (upper body link) is a model with inertia lb around mass point A1.
- the mass of each of the mass points A1 to A5 and the position of each of the mass points A1 to A5 in the local coordinate system fixed to the corresponding part are the same as those of the first displacement dimension correcting model.
- the inertia lb of the upper body 3 is the same as the first displacement dimension correcting model.
- the posture of the robot 1 is not restricted to the upright posture state as in the first displacement dimension correction model, and each of the mass points A1—A5 and the body 3 (top The body link) can be moved to a position and posture corresponding to any posture state that the robot 1 can take.
- Geometric constraint (2) The position and orientation of the part corresponding to each element of the robot 1, which is determined by the arrangement of each element of the second displacement dimension correction model, and the robot 1 in the instantaneous target movement corresponding to the arrangement. And the position and orientation of the part corresponding to each of the elements.
- the overall center of gravity force of the mass points A1 to A5 corresponds to the arrangement of each element of the second displacement dimension correction model (the position of the mass points A1 to A5 and the posture of the body link). The position almost coincides with the position of the true overall center of gravity of the robot 1 in the posture state.
- the arrangement force of the elements of the second displacement dimension correction model is determined by the arrangement of the parts corresponding to the elements (position This is equivalent to determining the arrangement of each element of the second displacement dimension correction model so as to match the posture.
- Determining the instantaneous target motion from the arbitrary arrangement of the elements of the second displacement dimension correction model in accordance with the geometric constraint condition (2) means that a part corresponding to each element of the robot 1 following the instantaneous target motion is determined. This is equivalent to determining the instantaneous target motion so that the arrangement (position and orientation) of the object coincides with the arrangement of the elements of the given second displacement dimension correction model.
- the second displacement dimension correcting model is a model for determining the displacement dimension corrected body position / posture in cooperation with the first displacement dimension correction model, and the displacement dimension corrected body position / posture In deciding, two types of arrangement of the elements of the second displacement dimension correction model are provisionally determined.
- the position of each foot mass point A4, A5 of the second displacement dimension correction model is determined to be a position corresponding to each foot position / posture of the simple Eich model gait.
- the positions of the upper body mass point A1 and the thigh mass points A2 and A3, and the posture angle of the upper body 3 (upper body link) are the center of gravity of the first displacement dimension correction model and the second displacement dimension correction model. , And the angular momentum product between these models described later. This will be described later in detail.
- each leg 2 of the robot 1 of the present embodiment has six degrees of freedom, if the positions and postures of both feet 22, 22 and the position and posture of the upper body 3 are determined, the legs of the robot 1 are determined.
- the overall posture of the bodies 2 and 2 (the position and posture of each part (each link) of each leg 2 and 2 of the robot 1) is uniquely determined. Therefore, if the positions of both foot mass points A4 and A5 and the body mass point A1 and the posture of the body 3 (body link) on the second displacement dimension correction model are determined, the thigh mass points A2 and A3 are correspondingly determined. Is determined subordinately.
- each foot mass point A4 of the second displacement dimension correction model , A5 is determined to be the position corresponding to each foot position / posture of the simple Eich model gait. Furthermore, the posture of the body link is determined to be the same as the body posture of the simplified model gait.
- the positions of the upper body mass point A1 and the thigh mass points A2 and A3 are determined by the first displacement dimension correction. Is determined so as to satisfy a predetermined condition relating to an angular momentum product described later between the use model and the second displacement dimension correction model. Details of this will be described later.
- the displacement dimension correction body position / posture is finally determined based on the above two types of arrangement of the elements of the second displacement dimension correction model.
- the "position" of each mass point or the "posture” of a link having inertia relating to the simple model and the first and second displacement dimension correcting models is not particularly specified. As far as possible, it means the position and orientation in the global coordinate system.
- the gait generator 100 performs a target gait for one step from the time when one leg 2 of the robot 1 lands and the force when the other leg 2 lands, by the processing of the flowchart described below.
- the desired gait for one step is sequentially generated in units of (the desired gait).
- the new desired gait to be generated is called a “current gait”.
- FIG. 9 is a structured flowchart showing the main routine processing of gait generator 100.
- various initialization operations such as initializing the time t to 0 in S010 are performed first. This process is performed when the gait generator 100 is activated. Next, the process proceeds to S014 via S012, and the gait generator 100 waits for a timer interrupt for each control cycle (the arithmetic processing cycle in the flowchart of FIG. 9). The control cycle is At. Thereafter, the processing from S014 to S032 is repeated for each control cycle At.
- the process proceeds to S016, where it is determined whether or not it is a gait switch. If the gait is a switch, the process proceeds to S018. If not, the process proceeds to S022.
- the “gait switching point” means the timing at which the generation of the current time's gait is started. It becomes the switch of the gait.
- the current time t is initialized to 0, and then proceed to S020 to determine the gait parameters of the current time's gait.
- the processing of S020 corresponds to the processing of the gait parameter determination unit 100a of FIG. 5 described above, and includes parameters for defining the desired foot position / posture trajectory, the desired arm / posture trajectory, the desired ZMP trajectory, and the desired floor reaction force vertical component trajectory. And the parameters that define the reference body posture trajectory, floor reaction force horizontal component tolerance, and ZMP tolerance Data is determined.
- the processing of S020 is, for example, a processing corresponding to S022-S030 in FIG. 13 of PCT International Publication WOZ 03/057427 / A1 (hereinafter referred to as Patent Document 1 and! ⁇ ⁇ ) previously proposed by the present applicant. This is performed in the same manner as in the above-mentioned publication. To summarize this, first, a normal gait as a virtual periodic gait to which the gait should be connected or asymptotic (a gait in which the gait of two steps of the robot 1 is one cycle) ) Is determined.
- the normal gait is based on periodicity conditions (the initial state of one cycle of the normal gait ( It is determined so as to satisfy the condition that the position and orientation of each part of the robot 1 and its change speed) and the state of the terminal end match.
- a gait meter that specifies a desired foot position / posture trajectory, a target arm posture trajectory, a desired ZMP trajectory, and a desired floor reaction force vertical component trajectory so that the current time's gait is connected to the normal gait or asymptotically approaches. Is determined.
- the gait parameters that define the desired foot position / posture trajectory are, for example, when generating the desired foot position / posture trajectory using the finite time settling filter proposed by the applicant of the present invention in Japanese Patent No. 3233450.
- the reference body posture is, for example, a vertical posture (a posture in which the inclination angle of the body 3 with respect to the vertical axis is 0).
- the gait parameter defining the floor reaction force horizontal component allowable range is the floor for full model correction determined in S030 of FIG. This corresponds to the parameter of the allowable range of the reaction force horizontal component.
- a dynamic model is used for creating a normal gait, and the dynamic model is used.
- the simple Eich model is used.
- the mass of both leg mass points of the dynamic model of Patent Document 1 is set to 0, And upper body This is equivalent to setting the inertia of the flywheel (flywheel inertia) to 0. Therefore, if the mass of both masses of the legs in the dynamic model of FIG.
- the floor reaction force horizontal component allowable range for the simplified model gait (this allowable range is In the present embodiment, the floor reaction force for the simplified model gait horizontal component allowable range is, for example, an infinite range of force or The floor reaction force horizontal component of the simplified model gait (or the normal gait) may be set to a wide range that always falls within the floor reaction force horizontal component allowable range. By doing so, the algorithm shown in Patent Document 1 can be applied to the processing of S020 of the present embodiment without any problem.
- the process proceeds to S022, where the instantaneous value of the current time's gait is determined.
- This process is a process executed by the target instantaneous value generation unit 100b and the simple gait model gait generation unit 100c in FIG. 5, and based on the gait parameters determined in S020, the instantaneous gait of the current gait is calculated. The value (the instantaneous value of the simple dani model gait) is determined.
- This process is more specifically equivalent to the process of S032 in FIG. 13 of Patent Document 1, and is performed in the same manner as in Patent Document 1.
- the instantaneous values of the desired foot position / posture, target ZMP, target arm posture, target floor reaction force vertical component, and reference body posture are calculated. Is determined, and based on the instantaneous values, the desired ZMP and the desired floor reaction force vertical component are satisfied on the simplified model (the resultant force of the inertial force generated by the motion of the robot 1 and gravity is satisfied).
- the momentary horizontal component of the moment acting around the target ZMP becomes 0, and the instantaneous value of the target body position and orientation is determined so that the translational force vertical component of the resultant balances the target floor reaction force vertical component.
- the instantaneous value of the desired body posture is the same as the instantaneous value of the reference body posture in the present embodiment.
- the desired body position vertical component is determined corresponding to the vertical position of the body mass point 3m of the simple body model obtained from the target floor reaction force vertical component and the above equation 01.
- the horizontal position of 3m of the upper body mass point of the simplified model is It is determined so that the left side of 03 is set to 0 (so that the floor reaction force moment horizontal component around the target ZMP becomes the force SO), and the target body position horizontal corresponding to the horizontal position of this body mass point 3m The components are determined.
- the floor reaction force horizontal component allowable range for the simplified model gait is used, but in the present embodiment, the process of S020 will be described.
- the floor reaction force horizontal component allowable range for the simple gait model gait is, for example, an infinite range, or the floor reaction force horizontal component of the simplified model gait is always the floor reaction force. It must be set so that it falls within the allowable range of the horizontal component.
- the target gait (the current gait) for which the instantaneous value is determined sequentially (for each control cycle At) is, in short, the target gait on the simple ridge model.
- the horizontal component of the moment generated around the target ZMP is zero when the resultant force of the inertia force and gravity at which the motion occurs is generated, and the gait is such that the translational force vertical component of the resultant force is balanced with the desired floor reaction force vertical component. is there.
- This displacement dimension gait correction subroutine relates to the core of the present invention, and will be described in detail below.
- the target gait generation processing using the simplified model has the advantage that the current time's gait (do not diverge !, the current time's gait) can be determined stably in real time. Low dynamic approximation accuracy. For this reason, in the embodiment of the present invention, a part of the gait (the desired body position / posture, the moment around the desired ZMP) is corrected by using a full model having higher dynamic accuracy than the simple model. . In this case, because the dynamic approximation of the simplified model gait is low, the nonlinearity of the full model is strong, and so on, the simplified model gait is input to the full model.
- the gait may not be properly corrected, and a gait may not be able to perform the continuous motion of the robot 1, which may cause a problem.
- the gait model is taken into account in the simple dani model.
- the influence of the change in the inertial force due to the bending motion of the knee joint becomes large, the dynamic approximation accuracy of the simplified model gait is reduced, and the above-described inconvenience is likely to occur.
- a dynamic model having multiple masses on each leg 2 can be constructed, and inertia (inertia) can be added to one or more links of the robot, such as the upper body.
- inertia inertia
- the nonlinearity of the simple model increases, leading to a normal gait.
- the gait parameter of the current time's gait (a gait parameter that can ensure the continuity of the motion of the robot 1)
- the dynamic equations including the target ZMP and the floor reaction force are calculated using the first and second displacement dimension correcting models.
- Geometric processing on the arrangement of elements of the first and second displacement dimension correction models that are not to be used position and orientation of links with mass points and inertia (processing on displacement of position and orientation dimensions )
- Only a part of the movement of the simple Eich model gait (specifically, the body position and posture) is corrected.
- the gait with higher dynamic accuracy than the simplified model gait more specifically, the translational force component of the actual inertial force generated by the robot 1 due to the motion of the gait and the resultant force of the gravitational force is obtained.
- the upper body posture of the simple gait model gait are corrected by focusing only on enhancing the dynamic accuracy, the upper body posture may be excessively changed.
- the upper body 3 of the robot 1 is generally heavier and has a larger inertia than other parts, if the robot 1 moves in a gait in which the body posture changes frequently, Excessive moment at the hip joint. As a result, an excessive load is applied to the hip joint actuator, or the hip joint portion and its vicinity vibrate radially, and the posture of the robot 1 is likely to be quickly unstable.
- an imaging device is mounted as a visual device on the head 4 of the robot 1 or the like, it is easy for the imaging device to oscillate, and it becomes difficult to accurately recognize the environment using the imaging device.
- the target step reaction force vertical component has to be zero or very small.
- the horizontal component more precisely, the component parallel to the floor
- the inertial force that can be generated by the translation of the upper body 3 of the robot 1 is limited, and the translation of the upper body 3 within the limit is limited. This is because it will be difficult to meet the target ZMP only by adjustment.
- the first displacement dimension correction is performed in consideration of the motion form of the robot 1 in the target gait and the friction state of the floor.
- the model and the second displacement dimension correction model we decided to correct the body position and posture of the simplified model gait so as to increase the dynamic accuracy while minimizing fluctuations in the body posture.
- the first and second displacement dimension correcting models used for correcting the body position and orientation of the simplified model gait usually have the same mass points corresponding to several parts of the robot 1 in the same manner. Or a mass point and a link with inertia (such as upper body 3).
- the two displacement dimension correcting models have more mass points than the simple model, or have inertia that the simple model does not have.
- both displacement dimension correction models in each embodiment of this specification have a mass point and inertia corresponding to the upper body 3 in order to correct the body position and orientation of the simplified model gait.
- Body link
- the instantaneous value (instantaneous motion) of the position and orientation of each part of the generated simplified model gait is The position of each mass point of the one displacement dimension correction model is determined.
- the first displacement dimension correction Model force If the model has inertia at one or more links (such as the upper body 3) of robot 1, the attitude angles of the links are also determined.
- appropriate geometrical constraints such as the positional relationship of each mass point or the posture of a link (such as the upper body 3) with inertia are added.
- the position of each mass of the first displacement dimension correction model and the attitude angle of the link with inertia are determined.
- a floor reaction force similar to the floor reaction force of the simplified model gait is generated in the first displacement dimension correcting model.
- the geometric constraint (1) is added to the first displacement dimension correcting model as such a geometric constraint.
- condition 1 is satisfied between the second displacement dimension correction model having no geometric constraint set in the first displacement dimension correction model and the first displacement dimension correction model.
- the position of each mass point of the second displacement dimension correction model that satisfies, 2 is the first element arrangement of the second displacement dimension correction model. Is provisionally determined as
- condition 1 is a condition in which the inertia force generated by the translational floor reaction force or the movement of the overall center of gravity is almost the same in the two displacement dimension correcting models.
- condition 1 is a vector of the difference (position vector difference) between the position of each mass point of the first displacement dimension correction model and the corresponding mass point position of the second displacement dimension correction model. Is equivalent to the condition that the sum of the translational force component (mass of mass point * translational acceleration) of the inertial force generated by each mass point for all the mass points when it is regarded as the translational acceleration of the mass point is almost zero. is there
- the angular momentum product relating to Condition 2 corresponds to each mass point of each displacement dimension correcting model.
- the reference position to be determined is arbitrarily determined and the position of the point Q is arbitrarily determined, the following is defined for each of the mass points.
- the reference attitude angle for each link having the inertia is determined. When each is arbitrarily defined, it is defined as follows for each link.
- the angular momentum product of the mass point of each displacement dimension correcting model is obtained by calculating the line segment (vector of the line segment) connecting the point Q and the point at the reference position corresponding to the mass point, and the The reference point is equivalent to the product of the cross product of the displacement of the force and the displacement (the displacement vector) and the mass of the mass point.
- a product having a proportional relationship with the product of the outer product and the mass or a product approximately equivalent to the product of the outer product and the mass may be defined as the angular momentum product related to the mass point.
- the angular momentum product of the link having inertia of each displacement dimension correcting model is obtained by calculating the difference between the attitude angle of the link from the reference attitude angle corresponding to the link and the inertia of the link. It is equivalent to the product.
- a product having a proportional relationship with the product of the deviation of the attitude angle of the link from the reference attitude angle and the inertia or that approximately equal to the product is defined as the angular momentum product of the link. You may.
- the angular momentum product related to a given mass point corresponds to a line segment connecting the mass point and the predetermined point, and the mass point. It becomes a function (monotonically increasing function or monotonically decreasing function) that changes monotonically with respect to the angle between the reference point and the line segment connecting the predetermined point.
- condition 2 is, more specifically, the position of each mass point of the first displacement dimension correction model corresponding to each mass point of the second displacement dimension correction model.
- the attitude angle of each link with inertia of the first displacement dimension correction model is the reference attitude angle of each link with inertia of the second displacement dimension correction model.
- the condition is that the sum of the angular momentum products of the second displacement dimension correction model becomes a certain value.
- condition 2 is a condition between the position of each mass point of the first displacement dimension correction model and the position of the corresponding mass point of the second displacement dimension correction model.
- the vector of the difference (difference between the position vectors) is regarded as the translational acceleration of the mass point
- the difference between the attitude angles of each link having inertia is regarded as the angular acceleration of the link
- the translational force component of the inertial force generated by each mass acts on the moment around the point Q and the inertia.
- This is equivalent to the condition that the sum of the moment of inertia of the link (the moment of inertia of the rotational motion) and the moment acting around point Q becomes a certain value (predetermined value).
- the point Q is set to, for example, a target ZMP.
- Point Q is not limited to the target ZMP, but this will be added later.
- the body posture of the robot 1 is restricted to the posture in the simple gait model gait for the second displacement dimension correcting model.
- the position of each mass point of the second displacement dimension correction model that satisfies the condition 2 above (or the position of each mass point and the attitude of the link having inertia) with the first displacement dimension correction model Angle) is provisionally determined as the second element arrangement of the second displacement dimension correcting model.
- weighting is performed between the body posture corresponding to the first element arrangement and the body posture corresponding to the second element arrangement by using weights according to the motion form in the target gait of the robot 1 and the friction state of the floor.
- the average is determined as the displacement dimension corrected body posture, and the weighted average of the body position corresponding to the first element arrangement and the body position corresponding to the second element arrangement is determined as the displacement dimension corrected body position.
- the robot 1 uses a weight corresponding to the motion form in the target gait and the friction state of the floor with respect to the second displacement dimension correction model.
- the second displacement dimension correction that satisfies the condition 2 above, while restricting the body posture of the simplified model gait to the weighted average of the body posture corresponding to the first element arrangement and the body posture of the simplified model gait
- the position of each mass point of the model for use (the position of each mass point and the attitude angle of the link having inertia) is determined as the second element arrangement of the second displacement dimension correction model.
- the body position / posture corresponding to the second element arrangement is directly determined as the displacement dimension corrected body position / posture.
- the target body position / posture of the simplified model gait is corrected using the first displacement dimension correction model and the second displacement dimension correction model as described above.
- the following displacement dimension corrected body position / posture is obtained.
- the process of S024 in the flowchart of FIG. 6 is a process of obtaining the displacement dimension corrected body position and orientation as described above.
- the subroutine processing of S024 in the first embodiment will be specifically described with reference to FIG.
- correction of the body position / posture (calculation of displacement dimension correction body position / posture) on a sagittal plane (a plane including the X axis and the Z axis) will be described.
- the explanation is omitted, and the correction of the body position / posture on the lateral plane (plane including the Y axis and the Z axis) is omitted.
- the first temporary dimension is set so as to satisfy Condition 1 regarding the center of gravity between the first displacement dimension correction model and the second displacement dimension correction model and Condition 2 regarding the angular momentum product.
- the corrected body position / posture (Pb21, 0b21) (a set of the first provisional corrected body position Pb21 and the first provisional corrected body posture ⁇ b21) is determined.
- the first element arrangement of the second displacement dimension correcting model is determined so as to satisfy the conditions 1 and 2 between the two models, and the position of the body mass point A1 in the first element arrangement is determined.
- the body position and posture of the robot 1 corresponding to the posture of the body link are determined as the first temporary corrected body position and posture (Pb21, ⁇ b21).
- the process of S100 is executed by a subroutine process of FIG.
- the first displacement dimension is calculated based on the instantaneous value of a simple gait model gait at this time (current time) t (instantaneous value of a target motion such as a desired body position / posture).
- the positions of the mass points A1 to A5 of the correction model and the attitude angle of the body 3 (body link) having inertia are obtained.
- the first displacement order is set so that the position of the entire center of gravity of the robot 1 in the simple dani model gait is equal to the position of the entire center of gravity of the robot 1 on the first displacement dimension correcting model.
- the positions of the mass points A1 to A5 of the original correction model are determined.
- the position of the overall center of gravity of the robot 1 in the simplified model gait coincides with the position of the upper body mass 3 m of the simplified model, and the position is the same as that of the simplified model gait. It is uniquely determined from the target body position.
- the positions of all the mass centers of the mass points A1 to A5 are uniquely determined by matching the position of the entire center of gravity of the robot 1 in the use model) with the position of the upper body mass point 3m of the simplified model.
- the posture angle of the body link of the first displacement dimension correcting model is the same as the body posture angle of the simplified model gait (vertical posture in this embodiment).
- the first displacement dimension correcting model is obtained from the instantaneous motion of the simple gait model gait (the instantaneous value at the current time t) according to the geometric constraint condition (1) relating to the first displacement dimension correcting model. Will be determined.
- the arrangement of the elements of the first displacement dimension correcting model corresponds to the “first arrangement” in the first invention of the present invention.
- the processing from S202 is executed, and the positions of the mass points A1 to A5 of the second displacement dimension correction model satisfying the above conditions 1 and 2 with respect to the first displacement dimension correction model, A set with the attitude angle of the body 3 (body link) having an inertia, that is, the first element arrangement of the second displacement dimension correcting model is exploratoryly determined, and the upper element in the first element arrangement is determined.
- the body position and posture of the robot 1 corresponding to the body constitution point A1 and the posture of the body link are determined as the first temporary corrected body position and posture (Pb21, 0b21).
- an initial candidate (Pb21_s, ⁇ b21_s) of the first temporarily corrected body position and orientation is determined.
- the initial candidates (Pb21_s, ⁇ b21_s) correspond to the approximate predicted values of the first temporarily corrected body position Pb21 and the first temporarily corrected body posture ⁇ b21 at the current time t (current time t). Is determined as follows, for example. That is, the difference between the first provisional corrected body position Pb21 at the current time t and the body position Pb of the simple dani model gait (the amount of displacement) is the previous time (the time of the previous control cycle) tA t It is considered to be close to the difference between Pb21 and Pb.
- the difference between the first temporary corrected body posture ⁇ b21 at the current time t and the simplified model gait body posture ⁇ b is ⁇ at the previous time t-At.
- the initial candidates Pb21_s, 0b21_s
- ⁇ b21 value Pb21_p, ⁇ b21_p and force are also determined by the following equations.
- Pb21_s Pb + (Pb21_p-Pb_p) ... Equation 04a
- ⁇ b21_s ⁇ b + ( ⁇ b21_p- ⁇ b_p)... Expression 05a
- the loop processing of S206-S216 is executed.
- the second displacement based on the current candidate of the first temporary corrected body position / posture (Pb21-s, ⁇ b21_s) and the target both foot position / posture of the simplified model gait at time t this time is used.
- Each mass point A1 in the dimensional correction model I Find the position of A5.
- the position and orientation of the upper body 3 of the robot 1 in the second displacement dimension correction model matches the current candidate (Pb21_s, ⁇ b21_s), and each of the robots 1 in the second displacement dimension correction model Assuming that the position and posture of the foot 22 match the target foot position and posture of the simplified model gait, the positions of the mass points A1 to A5 are obtained. In other words, of the instantaneous movements of the simplified model gait, only the instantaneous values of the body position and posture are replaced with candidates (Pb21_s, ⁇ b21_s), the geometric constraint condition ( According to 2), the positions of the mass points A1 to A5 in the second displacement dimension correcting model are obtained.
- the positions of the foot mass points A3 and A4 are determined from the desired foot position and orientation.
- the position of the body mass point A1 is determined from the candidate (Pb21_s, ⁇ b21_s), and the posture angle of the body 3 (body link) is the same as ⁇ b21_s.
- the position of each thigh mass point A2, A3 is determined from the posture of each leg 2 of the robot 1 determined from the target both foot position / posture and the candidate (Pb21_s, ⁇ b21_s).
- each leg 2 since each leg 2 has six degrees of freedom, if the positions and postures of both feet 22, 22 and the upper body 3 are determined, each leg 2 is determined.
- each part of the body 2 are also uniquely determined. Therefore, if the position of the upper body mass point A1, the posture angle of the upper body link, and the positions of both foot mass points A4 and A5 are determined, the positions of the thigh mass points A2 and A3 are uniquely determined. .
- Pil is a code generally indicating the position of mass point Ai of the first displacement dimension correction model
- Pi2 is generally indicating the position of mass point Ai of the second displacement dimension correction model.
- the posture angles of the body 3 (body link) in the first and second displacement dimension correcting models are generally represented by 0 bl and 0 b2, respectively.
- ⁇ bl is the same as the desired body posture 0 b (vertical posture) of the simplified model gait.
- Q is the same as the position of the target ZMP of the simplified model gait.
- This means the difference between the position of the center of gravity and the position of the overall center of gravity determined by the position Pi2 (i l, 2, ⁇ ⁇ ⁇ , 5) of each mass point A1-A5 of the second displacement dimension correction model. Therefore, if the value of the overall center-of-gravity deviation Gc_err between the models is 0 (0 vector) or almost 0, the above condition 1 is satisfied.
- the term excluding "Const” from the right-hand side of equation 07 means the sum of the angular momentum products of the second displacement dimension correction model with respect to the first displacement dimension correction model.
- (Pil-Q) * (Pi2—Pil) is a vector of a line segment connecting point Q and mass point Ai. And the displacement vector of the mass point Ai of the second displacement dimension correction model with respect to the mass point Ai of the first displacement dimension correction model.
- equation 07 relating to the angular momentum product
- equation 08-10 any of the following equations 08-10 may be used instead.
- the “angle (Pil_Q_Pi2)” in the equation 08 is a line segment connecting the mass point Ai and the point Q of the first displacement dimension correction model, and It means the angle between the line connecting the mass point Ai and the point Q of the second displacement dimension correction model.
- ⁇ Ci '' in Equation 08 is a predetermined coefficient, and its value is the area of a triangle formed by the mass point Ai and the point Q of the model for correcting Ci * mi * angle (Pil_Q_Pi2) force and two displacement dimensions.
- the “horizontal component displacement of mass point A i” in equations 09 and 10 is the horizontal displacement (Pi2—Pil) between mass point Ai of the first displacement dimension correction model and mass point Ai of the second displacement dimension correction model.
- “Height” means the relative height of the mass point Ai of the first or second displacement dimension correcting model with respect to the point Q, that is, the vertical component of Pil-Q or Pi2-Q.
- “C (height of mass Ai)” is the relative height (the vertical component of Pil—Q or Pi2—Q) of mass Ai of the first or second displacement dimension correction model with respect to point Q.
- the terms following ⁇ on the right side of the above equation 07-10 are a line segment connecting the mass point Ai and the point Q of the first displacement dimension correction model, and a term of the second displacement dimension correction model. It is a function that changes almost monotonically with the angle (Pil_Q_Pi2) formed by the line connecting the mass point Ai and the point Q.
- the positions of the mass points A1 to A5 of the first displacement dimension correction model obtained in S200 are substituted into Pil of the above equation (6), and the Pil of the above equation (6) is substituted.
- the center-of-gravity center position shift Gc_err between the models is calculated.
- Pil and Pi2 in the above equation (7) are made the same as equation (6), and the body posture (vertical posture in this embodiment) obtained in S200 is substituted into ⁇ bl, and 0b2 is further substituted into 0b2. 1
- the inter-model angular momentum is multiplied to calculate Lc_err.
- a Pb21x and A Pb21y are predetermined values for changing the current value of the candidate Pb2 l_s at the first provisional correction body position by a small amount in the X-axis direction and the Y-axis direction, respectively, and ⁇ 0b21 is the first provisional correction.
- This is a predetermined value for slightly changing the body posture candidate 0 b21 around the Y axis. Then, the same processing as in S206 and S208 is performed on each of these temporary candidates to obtain the overall center of gravity shift Gc_err between models and the angular momentum product shift err between models.
- the process of S214 is a process for observing the degree of change between Gc_err and err when the first temporary correction body position / posture candidate (Pb21_s, ⁇ b21_s) is changed from the current value.
- the first temporary correction body position / posture such that Gc_err and err fall within a predetermined range near 0 by the loop processing of S206-S216 in other words, the condition The first provisionally corrected body position and orientation that satisfies 1 and 2 are obtained by search.
- a set of the first temporary correction body position and orientation when the condition of S210 is satisfied and the position of each mass point obtained in S206 immediately before S210 corresponds to the first element arrangement. It will be.
- This first element arrangement corresponds to the “second arrangement” in the first invention of the present invention.
- the first provisional corrected gait is obtained by correcting only the desired body position and posture of the simplified model gait, and includes the desired gait such as the desired foot position and posture, the desired ZMP, and the desired floor reaction force vertical component.
- the other components of the gait are the same as the simple dani model gait.
- the motion of the first temporary correction gait is based on the geometric constraint condition (2) from the arrangement of the second displacement dimension correction model when the condition of S210 is satisfied (the first element arrangement). It is the same as the determined instantaneous target movement.
- the motion of the first temporarily corrected gait corresponds to the first temporarily corrected instantaneous target motion in the first invention of the present invention. Therefore, the processing in S100 constitutes the first temporary correction motion determining means in the first invention of the present invention.
- the body posture in the second displacement dimension correction model is made the same as the body posture in the instantaneous value (instantaneous value at time t) of the simple Eich model gait, and the first displacement dimension correction
- the second temporarily corrected body position and orientation (Pb22, ⁇ b22) (the second temporarily corrected body position Pb22 and the second temporarily corrected body position Pb22) so as to satisfy Condition 2 regarding the angular momentum product between the model and the second displacement dimension correcting model.
- Temporary correction body posture ⁇ b22 is determined. More precisely, the body posture in the second displacement dimension correction model is simplified.
- the second element arrangement of the second displacement dimension correcting model is determined so as to satisfy the condition 2 between the two models with the same body posture as that of the upper body mass point, and the upper body mass point in the second element arrangement is determined.
- the body position and posture of the robot 1 corresponding to the position of A1 and the posture of the body link are determined as the above-mentioned second temporary corrected body position and posture (Pb22, ⁇ b22).
- the second provisional corrected body posture ⁇ b22 is set to be the same as the body posture of the simplified model gait, so that the process of S102 substantially satisfies condition 2 It can be said that this is a process of determining the second temporary correction body position Pb22 so as to satisfy the condition.
- the process of S102 is executed by a subroutine process of FIG.
- the first displacement dimension is calculated based on the instantaneous value of the simple gait model gait at this time (current time) t (the instantaneous value of the target motion such as the target body position / posture).
- the positions of the mass points A1 to A5 of the correction model and the attitude angle of the body 3 (body link) having inertia are obtained.
- This process is the same as the process of S200 in FIG. 11, and the arrangement of each element of the first displacement dimension correcting model to be obtained is the same as that obtained by the process of S200. Therefore, if the arrangement of each element of the first displacement dimension correcting model obtained in S200 is used as it is in the subroutine processing of FIG. 12, the processing of S300 may be omitted!
- the processing from S302 is executed, and the positions of the mass points A1 to A5 of the second displacement dimension correction model that satisfy the above condition 2 with respect to the first displacement dimension correction model are determined by A pair with the posture angle of the body 3 (body link) having the shear, that is, the second element arrangement of the second displacement dimension correcting model is exploratively determined, and the upper body mass point in the second element arrangement is determined.
- the body position / posture of the robot 1 corresponding to A1 and the posture of the body link are determined as the second temporary corrected body position / posture (Pb22, ⁇ b22).
- the process from S302 is substantially a process of determining the second temporarily corrected body position Pb22 so as to satisfy the condition 2.
- initial candidates (Pb22_s, ⁇ b22_s) of the second temporary correction body position and orientation are determined. That is, of the initial candidates (Pb22_s, ⁇ b22_s), Pb22-s is defined as the body position Pb of the simplified model gait at the current time t and the simplified model gait at the previous time tAt. Body position value Pb_p and second temporary corrected body position at previous time t- ⁇ t The value Pb22_p and the force are also determined by the following equation 04b, and are also determined by the following equation 05b from the body posture ⁇ b of the simple model model gait at time t.
- the initial candidate Pb22_s of the second temporary correction body position is determined in the same manner as the initial candidate Pb21_s of the first temporary correction body position determined as described above in S202.
- the initial candidate ⁇ b22_s of the second temporary corrected body posture is the same as the body posture of the simplified model gait.
- This process is the same as S206 except for the value of the candidate (Pb22_s, ⁇ b22_s), and only the body position and orientation of the instantaneous motion of the simplified model gait at the current time t are candidates (Pb22_s, From the instantaneous motion replaced by ⁇ b22_s), the respective mass points A1 to A5 in the second displacement dimension correcting model are obtained according to the geometric constraint condition (2).
- a Pb22x and A Pb22y are predetermined values for changing the present value force of the first temporary correction body position candidate Pb22_s by a small amount in the X-axis direction and the Y-axis direction, respectively. Then, the same processing as in S306 and S308 is performed on each of these temporary candidates, and the angular momentum product shift err between the models is obtained. In the process of S314, the degree of change with respect to err when only the candidate of the body position is changed from the candidate of the second provisionally corrected body position and orientation (Pb22_s, ⁇ b22_s) is observed. It is processing for.
- the second temporary corrected body posture is set to be the same as the body posture of the simple gaiden model gait, and the L_err force ⁇
- a second provisionally corrected body position that falls within a predetermined range, in other words, a second provisionally corrected body position that satisfies the above condition 2 is obtained by search.
- the combination of the second provisional corrected body position and orientation when the condition of S310 is satisfied and the position of each mass obtained in S306 immediately before S310 corresponds to the second element arrangement. It becomes.
- the second element arrangement corresponds to the “third arrangement” in the first invention of the present invention.
- the process proceeds to S318 via S312, and the current b22_s, 0 b22_s) is determined as the second temporarily corrected body position and orientation (Pb22, 0 b22) at the current time t. Is done.
- a gait obtained by correcting the body position of the simple Eich model gait so as to satisfy the above condition 2 (hereinafter, also referred to as a second temporarily corrected gait) is obtained.
- the second temporary corrected gait is obtained by correcting only the desired body position in the simplified model gait, and includes the desired body posture, the desired foot position and posture, the desired ZMP, and the desired floor reaction force vertical.
- the other components of the target gait are the same as the simple gait model gait.
- the second provisional corrected gait is determined according to the geometric constraint condition (2) from the arrangement of the second displacement dimension correcting model when the condition of S310 is satisfied (the second element arrangement).
- Instantaneous goal Same as gait.
- the motion of the second temporarily corrected gait corresponds to the second temporarily corrected instantaneous target motion in the first invention of the present invention. Therefore, the processing in S102 constitutes the second provisional corrected motion determining means in the first invention of the present invention.
- a basic value wl_aim (hereinafter referred to as a weight basic value) relating to a weight wl for finally determining a displacement dimension corrected body position and orientation from the first temporary corrected body position and orientation and the second temporary corrected body position and orientation wl_aim).
- wl_aim l
- wl_aim 0.5
- wl_aim 0 is set.
- the value of the weight wl is gradually approached from the current value (the value determined at the previous time t At) to the weight basic value wl_aim determined as described above at S104 at the current time t.
- the current value of the weight wl is added. Determine the weight wl at time t. In this way, the weight wl is determined so as to gradually follow the basic weight value wl_aim with a response delay.
- weight w2 is determined so that the sum with the previously determined weight wl becomes 1. That is, the weight w2 is determined by the following equation 11.
- the displacement dimension corrected body position Pb2 is the sum of the first temporary corrected body position Pb21 and the second temporary corrected body position Pb22 obtained as described above multiplied by the weights wl and w2 at time t. In other words, it is obtained as a weighted average value of the first temporarily corrected body position Pb21 and the second temporarily corrected body position Pb22.
- the displacement dimension corrected body posture ⁇ b2 is the sum of the first provisional corrected body posture ⁇ b21 and the second provisional corrected body posture ⁇ b22 multiplied by weights wl and w2, in other words, the first The weighted average value of the temporary corrected body posture ⁇ b21 and the second temporary corrected body posture ⁇ b22 is obtained.
- the processing of the displacement dimension gait correction subroutine is executed, and the displacement dimension correction body position / posture is obtained.
- a desired gait (hereinafter, sometimes referred to as a displacement dimension corrected gait) obtained by correcting the body position and orientation of the simplified model gait can be obtained.
- This displacement dimension corrected gait is obtained by correcting only the desired body position and posture of the simplified model gait, and includes the desired foot position and posture, the desired ZMP, and the desired floor reaction force vertical component.
- the other components of the gait are the same as the simple dani model gait.
- the motion of the displacement dimension corrected gait in the first embodiment corresponds to the corrected instantaneous target motion in the first invention of the present invention. Therefore, the processing of the displacement dimension gait correction subroutine of S024 constitutes the target motion correcting means in the first invention.
- the second displacement dimension in the case of determination (that is, the case where the simple gait model gait is not corrected in S024)
- the position of each mass point Ai of the correction model and the posture angle of the body 3 (body link), and the position and height of each mass point Ai of the first displacement dimension correction model determined corresponding to the simplified model gait Illustrate the relationship with the posture angle of body 3 (body link)!
- the position of each mass point Ai and the posture angle of the body 3 of the second displacement dimension correction model that is, the arrangement of the elements of the second displacement dimension correction model, in other words, are simplified models. From the instantaneous motion of the gait, it can be said that the gait was determined in accordance with the geometric constraint (2).
- Fig. 15 shows the position of each material point Ai and the posture angle of the body 3 (body link) of the second displacement dimension correction model corresponding to the first provisional corrected gait, and the first displacement dimension.
- the relationship between the position of each mass point Ai of the correction model and the posture angle of the body 3 (body link) is illustrated.
- the arrangement of each element of the second displacement dimension correction model in FIG. 15 corresponds to the instantaneous value of the gait of the simple model model assumed in FIG. This is the first element arrangement of the second displacement dimension correcting model determined as follows. This arrangement is the same as that determined from the first temporary corrected gait according to the geometric constraint condition (2).
- the position of each mass point Ai and the posture angle of the body 3 (body link) of the first displacement dimension correcting model shown in FIG. 15 are the same as those in FIG.
- Fig. 16 shows the position of each material point Ai and the posture angle of the body 3 (body link) of the second displacement dimension correcting model corresponding to the second provisional corrected gait, and the first displacement dimension.
- the relationship between the position of each mass point Ai of the correction model and the attitude angle of the body 3 (body link) is illustrated.
- the arrangement of each element of the second displacement dimension correcting model in FIG. 16 corresponds to the instantaneous value of the simple gazing model gait assumed in FIG. This is the second element arrangement of the second displacement dimension correcting model determined as follows. This arrangement is the same as that determined according to the geometric constraint condition (2) from the second temporary corrected gait.
- the position of each mass point Ai and the posture angle of the body 3 (body link) of the first displacement dimension correcting model shown in FIG. 16 are the same as those in FIG. [0215]
- the displacement between an arbitrary mass point of the first displacement dimension correction model and the corresponding mass point of the second displacement dimension correction model when the displacement is regarded as translational acceleration is considered.
- the translational acceleration is referred to as a pseudo-translational acceleration between models of the mass point.
- the angular acceleration when the deviation of the attitude angle of the link with the inertia of the first displacement dimension correction model and the corresponding link of the second displacement dimension correction model is regarded as the angular acceleration is ,
- the arrangement of the elements of the second displacement dimension correcting model corresponding to the first provisional corrected gait (first element arrangement), that is, is finally determined in S100 of FIG.
- first element arrangement the arrangement of the elements of the second displacement dimension correction model determined by the position of each mass point Ai of the second displacement dimension correction model and the attitude angle of the body link, as shown in FIG. , 2 Mass point A2—
- Positional force of the body mass point Al of the model is determined in front of the body mass point Al, in other words, the first temporary corrected body position is ahead of the simplified model gait. Corrected to the side.
- the position of each mass point A1-A5 of the second displacement dimension correcting model and the posture of the body 3 having inertia are determined so that the sum of the angular momentum products becomes a certain value.
- the body posture (the posture of the upper body 3 shown by a solid line) in the second displacement dimension correction model is the body posture of the simple dani model gait (the posture of the upper body 3 shown by a broken line). ) Is inclined backward by an angle of 0 b21-0 b.
- the first provisional corrected gait compensates for the effect of the inertial force due to the movement of each leg, which is not considered in the simplified model gait, and sets the desired floor reaction force of the simplified model gait.
- the goal of the robot 1 is to generate a floor reaction force similar to (translational floor reaction force and floor reaction force moment, more precisely, the floor reaction force generated by the simple dani model gait).
- the dynamic accuracy between the motion of the first provisional corrected gait and the floor reaction force is the same as that between the motion of the simplified model gait and the floor reaction force. Accuracy will be higher than the target accuracy.
- the arrangement of the elements of the second displacement dimension correction model corresponding to the second provisional corrected gait (second element arrangement), that is, the second displacement finally determined in S102 of FIG.
- the body posture is simplified. While maintaining the instantaneous body posture of the model gait, the model for the second displacement dimension correction Position force of each mass point Al-A5 The sum of the angular momentum products described above is determined to be a certain value.
- the second provisional corrected gait maintains the target body posture of the robot 1 the same as the body posture of the simplified model gait, but does not consider the simple gait model gait.
- the target body position of the robot 1 was corrected so that the floor reaction force moment similar to the floor reaction force moment of the simple dani model gait was generated by compensating for the effect of the inertial force due to the leg motion. It will be.
- the dynamic accuracy between the motion of the second provisional corrected gait and the floor reaction force moment is higher than the dynamic accuracy between the motion of the simple gait model gait and the floor reaction force moment.
- the second temporary correction gait determining the second element arrangement of the second displacement dimension correction model
- the shift of the overall center of gravity between the first and second displacement dimension correction models is taken into account. Therefore, the dynamic accuracy between the motion of the second provisionally corrected gait and the translational floor reaction force is not necessarily the dynamic accuracy between the motion of the simplified model gait and the translational floor reaction force. It is not always higher than the biological accuracy.
- the body position / posture of the corrected gait that is, the displacement dimension corrected body position / posture is determined by Expressions 12 and 13.
- the operation mode of the robot 1 is steadily set to the normal mode (operation modes other than the running mode and the low-friction floor walking mode, in which the robot 1 walks on the normal floor having a high friction coefficient, etc.).
- the displacement dimension corrected body posture is the posture (base (Quasi-posture), specifically, in this embodiment, the vertical posture is maintained, and the displacement dimension correction body position is the same as the second provisional correction gait body position (second provisional correction body position).
- the displacement dimension corrected gait corrects the body position of the simplified model gait while keeping the body posture intact (the body posture is not corrected) and the motion of the displacement dimension corrected gait.
- This is a gait that can improve the dynamic accuracy between the floor reaction force moment and the simple gait model gait.
- the motion of the displacement dimension-corrected gait satisfies the target ZMP more accurately when the actual robot 1 is operated according to the motion than when the actual robot 1 is operated according to the motion of the simplified model gait. It will be.
- the displacement dimension corrected body posture and the displacement dimension corrected body position are respectively set to the first temporary. It becomes the same as the corrected gait upper body posture (first temporary corrected body posture). For this reason, the displacement dimension corrected gait corrects both the body position and body posture of the simplified model gait, and the movement of the displacement dimension corrected gait and the floor reaction force (translation floor reaction force and floor reaction force). This is a gait in which the dynamic accuracy between the gait and the force moment can be higher than that of the simple dani model gait.
- the displacement dimension corrected body posture is higher than the first provisionally corrected gait.
- 1 Z2 posture angle of the body posture angle (the body posture angle in which the tilt angle with respect to the vertical direction is smaller than the body posture angle of the first provisionally corrected gait, more precisely, the body of the first provisionally corrected gait 1Z2 of the posture angle and 1Z2 of the body posture angle of the second provisionally corrected gait)
- the displacement dimension corrected body position is the body position of the first provisionally corrected gait and the second provisionally corrected gait. And an intermediate position with the upper body position.
- the displacement dimension correction gait suppresses the fluctuation of the body posture, and the movement of the displacement dimension correction gait and the floor reaction force.
- Both the torso position and body posture of the simplified model gait are corrected so that the dynamic accuracy between (the translational floor reaction force and the floor reaction force moment) is higher than that of the simple gait model gait. It becomes.
- the displacement dimension corrected body position to an intermediate position between the body position of the first provisionally corrected gait and the body position of the second provisionally corrected gait, Therefore, it is possible to prevent the translational inertial force horizontal component generated in the robot 1 from becoming excessive.
- the displacement dimension corrected body posture is set to an inclination angle smaller than the body posture angle of the first provisionally corrected gait, the target ZMP can be satisfied while suppressing fluctuations in the body posture. . Therefore, when the operation mode of the robot 1 is constantly in the low-friction floor walking mode, the displacement dimension correction gait increases the dynamic accuracy more than the simple Eich model gait and suppresses the fluctuation of the body posture. A gait that enables stable operation of the robot 1 while minimizing and preventing the robot 1 from slipping.
- the instantaneous value of the ZMP allowable range and the instantaneous value of the floor reaction force horizontal component allowable range for the full model correction are determined. Is performed. This is a process executed by the target instantaneous value generation unit 100b, and is based on the gait parameters defining the ZMP allowable range and the floor reaction force horizontal component allowable range among the current time's gait parameters determined in S020. Is determined.
- the full model used by the full model correction unit 100e includes, for example, the upper body 3 of the robot 1, the hip joint of each leg 2, the thigh link, the lower leg link, the ankle joint, and the foot 2 as shown in FIG.
- This is a multi-mass model in which each has a mass point in 2 and an inertia lb in body 3 (body link).
- an inertia may be set for links other than the upper body 3.
- the operation of the composite compliance control apparatus 101 will be described with reference to FIG.
- the operation of the composite compliance control device 101 is described in detail in Japanese Patent Application Laid-Open No. Hei 10-277969 filed earlier by the present applicant, so that only a brief description will be given in this specification.
- the gait generator 100 among the target gaits generated as described above, the corrected target body position / posture (trajectory) and the target arm posture (trajectory) are converted into a robot geometric model (inverse kinematics calculation unit). Sent to 102.
- the desired foot position / posture (trajectory), the desired ZMP trajectory (target total floor reaction force center point trajectory), and the desired total floor reaction force (trajectory) (corrected desired floor reaction force moment and desired floor reaction force)
- the vertical component is sent to the composite compliance operation determining unit 104 and also to the target floor reaction force distributor 106.
- the desired floor reaction force distributor 106 distributes the floor reaction force to each foot 22, and determines the desired center of each foot floor reaction force and the desired foot floor reaction force.
- the determined target foot floor reaction force center points and the desired foot floor reaction force are sent to the composite compliance operation determination unit 104.
- the corrected target foot position / posture (trajectory) with mechanism deformation compensation is sent from the composite compliance operation determination unit 104 to the robot geometric model 102.
- the corrected target foot position / posture with the mechanism deformation compensation is such that the actual floor reaction force detected by the six-axis force sensor 50 approaches the target floor reaction force while taking into account the deformation of the compliance mechanism 72 of each leg 2. So each foot This means that the target foot position and orientation of 22 have been corrected.
- the robot geometric model 102 receives the 12 joints of the legs 2 and 2 that satisfy them. Calculate the joint displacement command (value) of the joint and send it to the displacement controller 108.
- the displacement controller 108 controls the displacement of the twelve joints of the robot 1 according to the joint displacement command (value) calculated by the robot geometric model 102 as a target value. Further, the robot geometric model 102 calculates a displacement designation (value) of the arm joint that satisfies the target arm posture, and sends it to the displacement controller 108. The displacement controller 108 controls the displacement of the twelve joints of the arm of the robot 1 with the joint displacement command (value) calculated by the robot geometric model 102 as a target value.
- the floor reaction force generated by the robot 1 (specifically, the actual floor reaction force of each foot) is detected by the 6-axis force sensor 50.
- the detected value is sent to the composite compliance operation determining unit 104.
- the posture tilt deviation ⁇ errx, erry erry generated at the mouth bot 1 (specifically, the deviation of the actual body posture angle from the target body posture angle, and the posture angle deviation in the roll direction (around the X axis) is ⁇ errx.
- the attitude angle deviation in the pitch direction (around the Y axis) is erryerry) is detected via the attitude sensor 54, and the detected value is sent to the attitude stabilization control calculation unit 112.
- the posture stabilization control calculation unit 112 calculates a compensation total floor reaction force moment around the desired total floor reaction force center point (target ZMP) for restoring the body posture angle of the robot 1 to the target body posture angle. This is sent to the composite compliance operation determination unit 104.
- the composite compliance operation determining unit 104 corrects the desired floor reaction force based on the input value. Specifically, the target total floor reaction force moment or the sum of the corrected total floor reaction force moment and the corrected target floor reaction force moment acts around the target total floor reaction force center point (target ZMP). Correct the floor reaction force.
- the composite compliance operation determination unit 104 adjusts the corrected target foot reaction force with the mechanical deformation compensation so that the corrected target floor reaction force matches the actual robot state and the floor reaction force for which the force such as a sensor detection value is also calculated. Determine the flat position and attitude (trajectory). However, since it is practically impossible to match all the states to the target, a trade-off relationship is given between them so that they can be compromised. In other words, the control deviation for each target is weighted, and control is performed so that the weighted average of the control deviation (or the square of the control deviation) is minimized. As a result, the actual foot position / posture and the total floor reaction force become the target foot position / posture and the desired total floor reaction force. Is generally controlled.
- the configurations of the robot 1 and the control unit 60 are the same as those of the first embodiment, and the simplified model, the first displacement dimension correcting model, and a part of the processing of the gait generator 100 are the same as those of the first embodiment. It is different from the form. Therefore, in the description of the present embodiment, the same portions as those in the first embodiment are denoted by the same reference numerals and drawings as those in the first embodiment, and detailed description thereof will be omitted.
- the second embodiment is an embodiment of the first, third, fourth, seventh to tenth, and thirteenth to 15th aspects of the present invention.
- FIG. 18 shows the structure of a simplified model (dynamic model) in the present embodiment
- FIG. 18 shows the structure of the first displacement dimension correcting model in the present embodiment.
- the simple model of the present embodiment shown in Fig. 18 has two foot masses 2m2, 2m2, and 2 corresponding to each leg 2 of the robot 1 (specifically, the foot 22 of each leg 2).
- This is a model composed of three masses consisting of a 3m2 upper body mass point corresponding to the upper body 3 and a flywheel FH having inertia J and no mass, which is the same as the model shown in FIG. is there. Accordingly, the power of omitting the detailed description in the present specification is as follows.
- the body mass point 3m2 is set to a point that is uniquely determined according to the position and orientation of the body 3 (a fixed point on the local coordinate system arbitrarily fixed to the body 3).
- the foot mass point 2m2 is set to a point uniquely determined in accordance with the position and orientation of the foot 22 of each leg 2 (a fixed point on the local coordinate system arbitrarily fixed to the foot 22). .
- the sum of the masses of the mass points 2m2, 2m2, 3m2 is the same as the total mass of the robot 1.
- the mass of the upper body mass 3m2 includes the mass of upper body 3 and the mass of both arms. It is.
- Equation 14-16 An equation (equation of motion) describing the dynamics of this simplified model is expressed by the following equation 14-16.
- equation 14-16 An equation of motion describing the dynamics of this simplified model is expressed by the following equation 14-16.
- the sagittal plane plane including the longitudinal axis (X axis) and the vertical axis (Z axis)
- the lateral plane left and right axes (Y axis)
- the equation of motion on the vertical axis the plane including the Z axis
- the variables in Equations 14 and 16 are defined as follows.
- Zsup Support leg foot mass point vertical position
- Zswg Swing leg foot mass point vertical position
- Zb Upper body mass point vertical position
- Xsup Support leg foot mass point horizontal position
- Xswg Swing leg foot mass point horizontal position
- Xb Upper body mass point horizontal position
- mb Upper body mass mass
- msup Support foot mass mass
- mswg Swing leg mass mass
- J Fly Wheel inertia moment
- Fx Floor reaction force horizontal component
- Fz Floor reaction force vertical component
- My Target Floor reaction force moment around the ZMP (more specifically, the component of the floor reaction force moment around the left and right axis (Y axis))
- Equation 16 In the second embodiment using the simplified model, a simple gait model gait that satisfies the target ZMP is generated in the same manner as in the above-mentioned Patent Document 1 as described later.
- the inertial force generated by the motion of the region near the knee joint due to the bending operation of the knee joint of each leg 2 is ignored. That is, the simplified model of the second embodiment is based on the knee joint associated with the bending operation of the knee joint of each leg 2. It can be said that this is a dynamic model constructed by assuming that the inertial force generated by the motion of the vicinity of is zero.
- the first displacement dimension correcting model according to the present embodiment corresponds to the upper body 3, the thigh link of each leg 2, and the foot 22.
- This is a 5-mass model consisting of upper body mass A1, thigh mass A2, A3, and foot mass A4, A5.
- the robot 1 also has an inertia (moment of inertia) lb around the body 3 (body link) force body mass point A1 of the robot 1.
- the first displacement dimension correcting model of the present embodiment includes, as in the first and second displacement dimension correcting models of the first embodiment, mass points A1 to A5 and a body link having inertia. It is configured as
- the upper body mass Al and each foot mass A4, A5 correspond to the corresponding part (the upper body 3, the foot, etc.), similarly to the first or second displacement dimension correcting model of the first embodiment. It is set to a point that is uniquely determined according to the position and posture of (Hei 22) (a fixed point in the local coordinate system arbitrarily fixed to the corresponding part). Note that the sum of the masses of the upper body mass Al, the foot masses A4 and A5, and the thigh masses A2 and A3 is equal to the total mass mtotal of the robot 1.
- the mass of the body mass point A1 includes the weight of the body 3 and the masses of the arms 5, 5 and the head 4.
- a certain geometric constraint condition is set for the arrangement of the elements of the first displacement dimension correcting model.
- the knee joint of each leg 2 of the robot 1 is directly expanded and contracted only in the direction connecting the center of the ankle joint of the leg 2 and the center of the hip joint. It is regarded as a dynamic (telescopic) joint, and each thigh mass point A2, A3 is set as an internal dividing point of a line connecting the center of the ankle joint of the leg 2 and the center of the hip joint.
- the internal dividing point is a point where the ratio of the distance from the internal dividing point to the center of the ankle joint and the distance from the center of the hip joint becomes a predetermined ratio, and each leg 2 is extended linearly.
- a point near the knee joint at that time (for example, a point slightly closer to the thigh link 24 than the center of the knee joint). Therefore, in the first displacement dimension correcting model in the present embodiment, the thigh mass points A2 and A3 are constrained to the inner dividing point of the line connecting the center of the corresponding ankle joint of the leg 2 and the center of the hip joint. !
- the thigh mass points A2 and A3 are separated by a predetermined distance in a direction perpendicular to the line segment from the inner dividing point. May be set to a point offset by only In other words, the thigh mass points A2 and A3 may be set on a straight line that is separated from the line by a predetermined distance and parallel to the line.
- the positions of the mass points A1 to A5 of the first displacement dimension correcting model on the global coordinate system and the attitude angle of the body 3 (body link) are calculated according to the simple model. It is determined geometrically according to the instantaneous value of the movement of the volume. More specifically, the position in the global coordinate system of the upper body mass A1 of the first displacement dimension correcting model of the present embodiment is determined to be a position corresponding to the body position / posture of the simplified model gait, The position of each foot mass point A4, A5 in the global coordinate system is determined to be a position corresponding to each foot position / posture of the simplified model gait. Furthermore, the posture angle of the body link is the same as the body posture of the simplified model gait.
- the positions of the thigh mass points A2 and A3 on the global coordinate system are determined to be the positions of the above-mentioned subdivision points determined based on the body position and posture of the simplified model gait and the position and posture of each foot.
- the position of the center point of each hip joint and each ankle joint of each leg 2 in the global coordinate system is uniquely determined according to the body position and posture and each foot position and posture of the robot 1.
- the positions of the thigh mass points A2 and A3 in the global coordinate system as internal dividing points of the line connecting the center point of the hip joint and the center point of the ankle joint of the leg 2 are determined.
- the overall center of gravity of the mass points A1 to A5 is the position of the overall center of gravity of the robot 1 on the simplified model, that is, all of the simplified models
- the predetermined ratio relating to the internal dividing point and the mass ratio of the mass points A1 to A5 are determined so as to match the positions of the centers of gravity of the mass points 2m2, 2m2, and 3m2.
- defining the arrangement of the elements of the first displacement dimension correction model as described above in the second embodiment is equivalent to the arrangement of the elements of the first displacement dimension correction model (in the global coordinate system).
- the geometric constraints (3) for defining the mass A1—the position of A5 and the posture of the body link) are defined as follows, the instantaneous motion of the simplified model gait Same as deciding the arrangement of the elements of the first displacement dimension correction model according to condition (3)
- Geometric constraints (3) For any given instantaneous target motion, the placement force of the upper body mass point A1 and the upper body link among the elements of the first displacement dimension correction model.
- the positions corresponding to the positions determined according to the position and orientation of the upper body 3 of the robot 1 in the target movement, and the positions of the foot mass points A4 and A5 correspond to the respective foot positions of the robot 1 in the given instantaneous target movement.
- the position of each thigh mass point A3, A4 coincides with the position determined in accordance with the posture, and the position of each thigh mass point A3, A4 is on the line connecting the center of the hip joint of each leg 2 and the center of the ankle joint in the given instantaneous target movement. It matches the position of the predetermined subdivision point.
- this geometric constraint (3) corresponds to the first geometric constraint in the present invention.
- the structure of the second displacement dimension correcting model is the same as that of the first embodiment shown in FIG. 1 Similar to the displacement dimension correction model, it has an upper body mass Al, thigh masses A2, A3, and foot masses A4, A5, and has an inertia lb in the upper body link.
- the positions of upper body mass point A1, each foot mass point A4, and A5 in the local coordinate system fixed to the corresponding part (upper body 3, each foot 22) are the first displacement dimension in Fig. 19. It is the same as the correction model.
- the mass of each mass point A1-A5 is the same as that of the first displacement dimension correcting model in FIG.
- each of the mass points A1 to A5 and the body 3 can be moved to a position and orientation corresponding to any posture state that the robot 1 can take. That is, the geometric constraint condition (2) described in the first embodiment is set between the arbitrary instantaneous target motion of the robot 1 and the arrangement of each element of the second displacement dimension correcting model. ing.
- the basic processing procedure of the gait generator 100 in the present embodiment is the same as in the first embodiment, and a gait is generated according to the flowchart of FIG.
- the process of S020 is performed after S018, and the gait parameters of the current time's gait are determined. That is, the parameters that define the desired foot position / posture trajectory, the desired arm posture trajectory, the desired ZMP trajectory, and the desired floor reaction force vertical component trajectory of the current time's gait are determined, The parameters that define the reference body posture trajectory, the floor reaction force horizontal component allowable range, and the ZMP allowable range are determined.
- the simplified model in the present embodiment is the same as the dynamic model used in the above-mentioned publication 1 as described above, and thus is the same as the process of S022-S030 in FIG.
- the gait parameters of the current time's gait are determined by executing the processing in S020 of the present embodiment.
- the floor reaction force horizontal component allowable range for the simple gait model gait is set and used in order to create a normal gait.
- the allowable range of the floor reaction force horizontal component for the simple gait model gait is, for example, the floor reaction force horizontal component for full model correction set in S30 of FIG.
- the force may be set to be the same as the component allowable range, or may be set to a wider range.
- the floor reaction force horizontal component allowable range for the simplified model gait is set to an infinite range, or the simplified model gait (or The gait) may be set to a wide range such that the floor reaction force horizontal component of the gait always falls within the floor reaction force horizontal component allowable range.
- the process proceeds to S022, where the gait parameters (gait parameters determined in S020) are used.
- the instantaneous value of the gait (simple Eich model gait) is determined.
- the simplified model in the present embodiment is the same as the dynamic model used in the above-mentioned publication 1, the same processing as that of S032 in FIG. 13 of the publication 1 is performed.
- the instantaneous value of the simple model model gait is determined.
- the instantaneous values of the desired foot position / posture, target ZMP, target arm posture, desired floor reaction force vertical component, and reference body posture are determined. Further, on the simple model shown in FIG. 16, the resultant force of the inertia force and the gravity force that generates the motion of the robot 1 becomes the moment horizontal component force ⁇ acting around the target ZMP, and The instantaneous value of the target body position and orientation is determined so that the force horizontal component does not exceed the floor reaction force horizontal component allowable range for the simplified model gait.
- the desired body position vertical component is the vertical position of the body mass point 3m2 of the simple dani model obtained from the desired floor reaction force vertical component and Equation 14 above. Is determined in correspondence with And During the period when the vertical component of the desired floor reaction force is relatively large, the instantaneous value of the target body posture is adjusted while adjusting the horizontal acceleration of the body 3 so that the moment horizontal component around the target ZMP becomes zero.
- the target body posture and the target body position horizontal component are determined so as to approach the reference body posture (for example, vertical posture).
- the horizontal acceleration of the body 3 is almost zero (strictly speaking, the horizontal acceleration of the entire center of gravity is almost zero).
- the instantaneous values of the desired body posture and the desired body position horizontal component are determined mainly by adjusting the angular acceleration of the posture angle of the body 3 so that the moment horizontal component around the target ZMP becomes zero.
- This subroutine process has the same basic processing procedure as that of the first embodiment, and is executed according to the flowchart of FIG. That is, first, in S100, as in the first embodiment, the condition 1 relating to the center of gravity between the first displacement dimension correcting model and the second displacement dimension correcting model, and the condition 2 relating to the angular momentum product
- the first temporary correction body position / posture (Pb21, ⁇ b21) is determined so as to satisfy the following.
- This processing is executed by the subroutine processing of FIG. 11, as in the first embodiment.
- the first displacement based on the instantaneous value of the simplified model gait at this time (current time) t (the instantaneous value of the target motion such as the target body position and orientation), the first displacement The positions of the mass points A1 to A5 of the model for dimension correction and the attitude angle of the body 3 (body link) with inertia are obtained.
- the position of the body mass point A1 of the first displacement dimension correction model is determined to be a position corresponding to the instantaneous value of the body position and posture of the simple Eich model gait
- each foot mass point A4 , A5 in the global coordinate system is determined to be the position corresponding to each foot position and posture of the simplified model gait.
- the positions of the thigh mass points A2 and A3 are determined by the center point of the hip joint and the center of the ankle joint of each leg 2 of the robot 1 determined based on the upper body position and posture and each foot position and posture of the simple gait model gait.
- a line segment connecting the point is determined at a position of an internally dividing point which is internally divided at a predetermined ratio.
- the posture angle of the body link of the first displacement dimension correcting model is the same as the body posture angle of the simplified model gait.
- the instantaneous motion of the simplified model gait (the instantaneous value at the current time t) is calculated according to the geometric constraint condition (3) according to the first displacement dimension correcting model in the present embodiment.
- the arrangement of each element of the displacement dimension correction model is determined.
- the arrangement of the elements of the first displacement dimension correcting model corresponds to the “first arrangement” in the first invention of the present invention.
- the processing from S202 to S218 is executed.
- These processes are the same as in the first embodiment. That is, the first temporary corrected body position / posture that satisfies the above conditions 1 and 2 is searched for and determined as the first temporary corrected body position / posture (Pb21, ⁇ b21) at the current time t. Is done.
- a gait (first temporary corrected gait) obtained by correcting the body position and orientation of the simplified model gait so as to satisfy the above conditions 1 and 2 is obtained.
- the motion of the first temporary correction gait is determined according to the geometric constraint condition (2) from the arrangement of the second displacement dimension correction model (the first element arrangement) when the condition of S210 is satisfied. It is the same as the instantaneous target exercise.
- the positions of both foot mass points A5 and A6 of each displacement dimension correction model are the same in both displacement dimension correction models. Therefore, in calculating the overall center of gravity shift Gc_err between models and the angular momentum product shift err between models in S208, the term relating to both foot mass points A5 and A6 may be omitted.
- the processing of S102 is executed in the same manner as in the first embodiment, and the body posture in the second displacement dimension correcting model is reduced to a simplified model gait.
- the condition 2 regarding the angular momentum product between the first displacement dimension correction model and the second displacement dimension correction model is satisfied by setting the same as the body posture at the instantaneous value (the instantaneous value at the current time t).
- the second temporary correction body position / posture (Pb22, ⁇ b22) is determined to be added.
- the second temporary corrected body posture ⁇ b22 is set to be the same as the body posture of the simplified model gait, so that the process of S102 substantially satisfies condition 2. It can be said that this is the process of determining the second temporary correction body position Pb22.
- This processing is executed by the subroutine processing of FIG. 12, as in the first embodiment.
- a second temporary corrected body position / posture in which the body posture is the same as the body posture of the simplified model gait and satisfies the condition 2 is exploratively obtained, and this is obtained at the current time t. It is determined as the second temporary correction body position / posture (Pb22, ⁇ b22).
- a gait (second temporary corrected gait) obtained by correcting only the upper body position of the simplified model gait so as to satisfy Condition 2 is obtained.
- the second provisional corrected gait is determined according to the geometric constraint condition (2) from the arrangement of the second displacement dimension correcting model when the condition of S310 is satisfied (the second element arrangement). It is the same as the instantaneous target gait.
- the positions of both foot mass points A5 and A6 of each displacement dimension correction model are the same in both displacement dimension correction models.
- the terms relating to the both feet mass points A5 and A6 may be omitted.
- the operation of the composite compliance control device 101 for inputting the desired gait generated as described above in the gait generating device 100 is the same as that in the first embodiment.
- processing of the displacement dimension gait correction subroutine in the second embodiment is described. Constitutes the target motion correcting means in the first invention of the present invention, and the motion of the displacement dimension corrected gait determined in this processing corresponds to the corrected instantaneous target motion in the first invention. Further, the processing of S100 and the processing of S102 in the second embodiment respectively correspond to the first temporary corrected motion determining means and the second temporary corrected motion determining means in the first invention, and the first temporary determined by the respective processing. The corrected gait and the second provisionally corrected gait correspond to the first provisionally corrected instantaneous target movement and the second provisionally corrected instantaneous target movement, respectively, in the first invention.
- the position of each mass point Ai and the posture angle of the body 3 (body link) of the second displacement dimension correction model The relationship between the position of each mass point Ai and the posture angle of the body 3 (body link) of the first displacement dimension correcting model determined corresponding to the gait of the simplified model!
- the arrangement of the elements of the second displacement dimension correcting model can be said to be determined from the instantaneous motion of the simplified model gait according to the geometric constraint condition (2).
- FIG. 21 shows the position of each material point Ai and the posture angle of the body 3 (body link) of the second displacement dimension correcting model corresponding to the first provisional corrected gait, and the first displacement dimension.
- the relationship between the position of each mass point Ai of the correction model and the posture angle of the body 3 (body link) will be illustrated.
- the arrangement of each element of the second displacement dimension correcting model in FIG. 21 corresponds to the instantaneous value of the simple gait model gait assumed in FIG. 20, and is finally determined in S100 in FIG.
- FIG. 21 shows the position of each material point Ai and the posture angle of the body 3 (body link) of the second displacement dimension correcting model corresponding to the first provisional corrected gait, and the first displacement dimension.
- the position of each mass point Ai and the posture angle of the body 3 (body link) of the first displacement dimension correcting model shown in FIG. 21 are the same as those in FIG. [0276]
- Fig. 22 shows the position of each particle Ai and the posture angle of the body 3 (body link) of the second displacement dimension correcting model corresponding to the second provisional corrected gait, and the first displacement dimension.
- the relationship between the position of each mass point Ai of the correction model and the posture angle of the body 3 (body link) will be illustrated.
- the moment in which the inertial force generated by each thigh mass A2, A3 due to the pseudo-translational acceleration between the thigh masses A2, A3 between the models acts around the target ZMP is generated on the backward tilt side of the robot 1. .
- the arrangement of the elements of the second displacement dimension correcting model corresponding to the first provisional corrected gait shows the arrangement of the elements of the second displacement dimension correction model determined by the position of each mass point Ai of the second displacement dimension correction model determined by
- first element arrangement that is, the final arrangement in S100 of FIG. Fig. 21
- the position force of the upper mass point A1 of the second displacement dimension correction model is higher than that of the upper mass point A1 of the first displacement dimension correction model.
- the body position is corrected to the rear of the simplified model gait.
- the positions of the mass points A1 to A5 of the second displacement dimension correcting model and the posture of the body 3 having inertia are determined so that the above-mentioned momentum products have a certain value (predetermined value).
- the body posture (the posture of the body 3 shown by a solid line) in the second displacement dimension correction model is the same as the body posture (the posture of the body 3 shown by a broken line) of the gait model. In contrast, it is inclined forward by an angle ⁇ b21-6 b.
- twice the area of the triangles with diagonal lines or horizontal lines in Fig. 21 correspond to the angular momentum products of the upper body mass Al and the thigh masses A2 and A3, respectively.
- the first provisional corrected gait compensates for the effect of the inertial force accompanying the motion of the portion near the knee joint of each leg, which is not taken into account in the simplified model gait.
- the target motion of robot 1 (more specifically, target body position and target body posture) is It will be corrected.
- the dynamic accuracy between the motion of the first provisional corrected gait and the floor reaction force is equal to the power between the motion of the simple dani model gait and the floor reaction force. It will be higher than the biological accuracy.
- the arrangement of the elements of the second displacement dimension correcting model corresponding to the second provisional corrected gait (second element arrangement), that is, finally determined in S 102 of FIG. 10 in the present embodiment. Is determined by the position of each mass point Ai of the second displacement dimension correction model and the posture angle of the body link.
- the second displacement Positional force of each of the mass points A1 to A5 of the dimension correction model is determined so that the sum of the angular momentum products described above becomes a certain value (predetermined value).
- each mass point Ai (more specifically, the upper body mass points A1 and A1
- the sum of the moments in which the inertial force due to the pseudo-translational acceleration between the thigh mass points A2 and A3) acts around the target ZMP is a predetermined value (corresponding to the ⁇ constant value '' in Condition 2 above) as shown in FIG. Value).
- the second provisional corrected gait maintains the target body posture of the robot 1 the same as the body posture of the simplified model gait, and does not take into account the simple gait model gait. Compensate for the effect of the inertial force due to the movement of the leg near the knee joint, and generate a floor reaction force moment similar to the floor reaction force moment of the simplified model gait. The body position is corrected.
- the dynamic accuracy between the motion of the second provisionally corrected gait and the floor reaction force moment is higher than the dynamic accuracy between the motion of the simple model model gait and the floor reaction force moment.
- the second provisional corrected gait determining the second element arrangement of the second displacement dimension correcting model
- the deviation of the overall center of gravity between the first and second displacement dimension correcting models is taken into account. Therefore, the dynamic accuracy between the motion of the second provisionally corrected gait and the translational floor reaction force is not necessarily the dynamic accuracy between the motion of the simple gait model gait and the translational floor reaction force. Accuracy is not always higher.
- Equation 13 Therefore, as in the first embodiment, when the operation mode of the robot 1 is constantly in the normal mode, the displacement dimension correction gait does not change the body posture (the body posture The body position of the simplified model gait is corrected while maintaining the dynamic accuracy between the motion of the displacement dimension corrected gait and the floor reaction force moment. It is a gait that can be enhanced.
- the displacement dimension correction gait is the same as the body position and the state / posture of the simplified model gait, as in the first embodiment. And the dynamic accuracy between the motion of the displacement dimension corrected gait and the floor reaction force (translational floor reaction force and floor reaction force moment) can be increased more than that of the simple gait model gait. Gait.
- the displacement dimension-corrected gait is more dynamically accurate than the simplified model gait, as in the first embodiment.
- the gait enables the robot 1 to perform a stable motion while increasing the body weight, minimizing the fluctuation of the body posture, and preventing the robot 1 from slipping.
- the configuration of the robot 1 is the same as that of the first and second embodiments, and the structures of the simple model, the first displacement dimension correcting model, and the second displacement dimension correcting model are the same as those of the second embodiment.
- the same as This embodiment differs from the second embodiment only in part of the processing of the gait generator 100. Therefore, in the description of the present embodiment, the same portions as those in the second embodiment will be denoted by the same reference numerals and drawings as those in the second embodiment, and will not be described in detail.
- the present embodiment will be described with a focus on portions different from the second embodiment.
- This embodiment is one embodiment of the second to fourth, seventh to tenth, and thirteenth to 15th inventions of the present invention.
- the processing of the gait generator 100 is different from that of the second embodiment only in the displacement dimension gait correction subroutine of S024 in FIG. Therefore, this displacement dimension gait Descriptions of processes other than the correction subroutine will be omitted.
- the displacement dimension gait correction subroutine in the present embodiment is performed as shown in the flowchart of FIG.
- the first and second displacement dimensions are corrected so as to satisfy Condition 1 regarding the center of gravity between the model for correcting the first displacement dimension and the model for correcting the second displacement dimension, and Condition 2 regarding the angular momentum product.
- 1Temporary correction body position and posture (Pb21, ⁇ b21) is determined.
- This processing is the same as the processing in S100 (FIG. 10) in the second embodiment, and is executed in exactly the same way as in the second embodiment by the subroutine processing in FIG. Supplementally, in the present embodiment (third embodiment), the arrangement of each element of the first displacement dimension correcting model obtained in the processing of S200 in FIG. 11 is the “first arrangement” in the second invention of the present invention.
- the arrangement of each element of the second displacement dimension correction model when the condition of S210 is satisfied corresponds to the first element arrangement, and the ⁇ second arrangement '' in the second invention of the present invention. ".
- the body posture in the second displacement dimension correction model is calculated by multiplying the first temporary correction body posture ⁇ b21 determined in S500 by the weight wl determined in S504. The same as the sum (wl * 0b21 + w2 * 0b) of the posture obtained by multiplying the body posture ⁇ b at the instantaneous value of the simplified model gait by the weight w2 determined in S506,
- the second provisionally corrected body position / posture (Pb22, ⁇ b22) is determined so as to satisfy Condition 2 regarding the angular momentum product between the one displacement dimension correcting model and the second displacement dimension correcting model.
- the second provisional corrected body posture ⁇ b22 is set to be equal to wl * 0b21 + w2 * ⁇ b, so that the process of S508 substantially satisfies condition 2 It can be said that this is a process of determining the second temporarily corrected body position P b22 so as to satisfy the condition.
- the process of S508 is executed by a subroutine process of FIG.
- This subroutine process is different from the subroutine process of FIG. 12 in the second embodiment only in the value of the second temporary correction body posture candidate ⁇ b22_s. That is, in the subroutine processing of FIG. 24 of the present embodiment (third embodiment), the candidate of the second temporary correction body posture b22-s is wl * 0 b21 + w2 * 0 b (The weight wl of the current time t is multiplied by the first provisional corrected body posture ⁇ ⁇ b21 of the current time t ⁇ ⁇ The weight w2 of the current time t is the current time 12 is different from the subroutine processing of FIG.
- Candidate ⁇ b22_s is fixed to the body posture at the instantaneous value of the simple-model model gait). More specifically, the subroutine process of FIG. 24 is the same as the subroutine process of FIG. 12 except that the initial candidate ⁇ b22_s is set to wl * ⁇ b21 + w2 * ⁇ b in the process of S602. . In this case, S600, S604-S618 force S in FIG. 24 are the same as S300, S304-S318 in FIG. 12, respectively.
- the arrangement of the elements of the second displacement dimension correction model when the condition of S610 is satisfied corresponds to the second element arrangement, and the second embodiment of the present invention. This corresponds to the “third arrangement” in the invention.
- the processing of the displacement dimension gait correction subroutine corresponds to the target motion correcting means in the second invention of the present invention, and the processing of the displacement dimension corrected gait determined by this processing is performed.
- the exercise in the third embodiment, this is equivalent to the exercise of the second provisionally corrected gait
- the processing in S500 corresponds to the provisionally corrected motion determining means in the second invention
- the first provisionally corrected gait determined in this processing corresponds to the provisionally corrected instantaneous target movement in the second invention.
- the first provisionally corrected gait is the same as in the second embodiment. Obedience
- the first provisional corrected gait as described with reference to FIGS. 20 and 21 is the inertial force due to the movement of the part near the knee joint of each leg, which is not considered in the simplified model gait.
- the robot 1 so as to generate a floor reaction force similar to the desired floor reaction force (translational floor reaction force and floor reaction force moment) of the simplified model gait. (Target body position and target body posture).
- the first provisional corrected gait is such that the actual floor reaction force generated when the actual robot 1 is operated according to the motion of the gait is the same as that of the simple robot model gait.
- Fig. 25 shows the arrangement (second element arrangement) of the elements of the second displacement dimension correcting model corresponding to the second provisional corrected gait in the present embodiment (third embodiment) and the first element. It illustrates the relationship with the arrangement of elements of the displacement dimension correcting model.
- the arrangement of each element of the second displacement dimension correction model in FIG. 25 corresponds to the instantaneous value of the gait model of the simple model shown in FIG. 20, and is finally determined in S506 of FIG.
- the second element arrangement of the second displacement dimension correcting model determined as follows. This arrangement is the same as that determined from the second provisional corrected gait according to the geometric constraint condition (2).
- Fig. 25 shows the arrangement (second element arrangement) of the elements of the second displacement dimension correcting model corresponding to the second provisional corrected gait in the present embodiment (third embodiment) and the first element. It illustrates the relationship with the arrangement of elements of the displacement dimension correcting model.
- the arrangement of each element of the second displacement dimension correction model in FIG. 25 corresponds to the instantaneous value of the
- the position of each mass point Ai and the posture angle of the body link of the first displacement dimension correcting model shown in FIG. 25 are the same as those in FIG.
- the body posture is changed to the first temporary corrected body posture ⁇ b21.
- the second displacement dimension correction is performed in a state where the posture angle obtained by multiplying the weight wl and the posture angle obtained by multiplying the weight w2 by the body posture ⁇ b at the instantaneous value of the simplified model gait is regulated.
- Positional force of each mass point A1-A5 of the model The sum of the angular momentum products is determined to be a certain value (predetermined value).
- each mass point Ai (more specifically, the upper body mass points A1 and The sum of the moments in which the inertial force due to the pseudo-translational acceleration between the thigh mass points A2 and A3) acts around the target ZMP is a predetermined value (corresponding to the ⁇ constant value '' in Condition 2 above) as shown in FIG. Value).
- the second provisional corrected gait is obtained by converting the target body posture of the robot 1 into the body posture (vertical posture in the present embodiment) at the instantaneous value of the simplified model gait. While compensating for the effect of the inertial force associated with the movement of each leg near the knee joint, which is not considered in the simple Eich model gait, The target body position of the robot 1 is corrected so that a floor reaction force moment similar to that of one robot is generated. In such a second provisional corrected gait, the actual floor reaction force moment generated when the actual mouth bot 1 is operated in accordance with the motion of the gait is changed according to the actual movement of the simplified model gait.
- the second provisionally corrected body position / posture is determined as the displacement dimension corrected body position / posture.
- the body position of the simplified model gait is corrected while the body posture is not fluctuated (maintained constant), and the power between the motion of the displacement dimension corrected gait and the floor reaction force moment is corrected. It is a gait that can improve the mechanical accuracy than the simple dani model gait.
- the simplified model gait is the same as in the second embodiment
- the displacement dimension corrected gait in the present embodiment (third embodiment) is also the same as in the second embodiment.
- the displacement dimension corrected body posture is higher than the first provisionally corrected gait. It is the sum of the 1Z2 posture angle of the body posture angle and the 1Z2 posture angle of the simplified model gait's body posture angle. Is the body position of the second provisionally corrected gait corresponding to the sum of the posture angle of the body posture angle of 1Z2 and the posture angle of the simplified model gait's body posture angle of 1Z2. You.
- the displacement dimension corrected gait in this case suppresses the movement of the displacement dimension corrected gait and the floor reaction force (translation floor reaction force and floor reaction force) while suppressing fluctuations in the body posture.
- Both the upper body position and upper body posture of the simplified model gait are corrected so that the dynamic accuracy between the gait and the simplified model gait is higher than that of the simplified model gait.
- the displacement-dimension-corrected gait has a higher dynamic accuracy than the simple-model model gait, suppresses fluctuations in the body posture as much as possible, and prevents the robot 1 from slipping.
- the displacement dimension corrected body position is determined so as to satisfy the above condition 2 in accordance with the displacement dimension corrected body posture as the corrected body posture. Therefore, the error between the actual floor reaction force moment and the desired floor reaction force moment generated when the actual robot 1 is operated in accordance with the movement of the displacement dimension corrected gait including the displacement dimension corrected body position / posture is calculated. It can be reduced effectively.
- the value of the weight wl is gradually changed by the processing of S504 in FIG. 23, and therefore, as in the first and second embodiments, the displacement dimension is changed.
- the corrected body posture and the displacement dimension corrected body position do not change suddenly.
- the displacement dimension corrected body posture is the first temporary corrected body posture. It is determined according to the power multiplied by the weight wl.
- the weight wl may have a frequency characteristic with respect to the first temporarily corrected body posture (the body inclination angle of the first temporarily corrected gait). For example, for the frequency component of the first temporary correction body posture, a low cut characteristic is given to the weight wl as shown in FIG. In such a case, a state in which the first provisionally corrected body posture is constantly and substantially kept constant, for example, a desired gait in which the robot 1 is continuously stopped in the upright posture state is set.
- the displacement dimension corrected body posture obtained by multiplying the first temporary corrected body posture by the weight wl can be reliably and constantly maintained in the vertical posture without causing an offset in the vertical direction. . For this reason, the appearance of the overall posture of the robot 1 is improved.
- the weight wl may have a high cut characteristic as shown in FIG. 26 (b).
- the displacement dimension correction body posture can be determined by removing high-frequency components of the first temporary correction body posture, that is, components that cause the first temporary correction body posture to vibrate at high speed. Become. As a result, fine vibrations of the displacement dimension correction body posture can be prevented from occurring, and the imaging device mounted on the head or the like of the robot 1 can be prevented from shaking.
- the target ZMP is used as the point Q related to the angular momentum product.
- the point Q may be a point other than the target ZMP. Good.
- the center of gravity of a certain set of mass points relating to the first and second displacement dimension correction models (specifically, there is a possibility that a difference in position may occur between the first and second displacement dimension correction models).
- the center of gravity of the set of all the mass points A1 to A5 is equivalent
- the centroid of the upper body mass point A1, the thigh mass points A2, and A3 is used.
- the center of gravity of the set is equivalent
- the displacement dimension gait correction subroutine of S024 in FIG. Force Performed by Process 10 As in the third embodiment, the force may be performed by the process of FIG.
- another embodiment according to the second invention of the present invention is configured.
- the body position of the simplified model gait on the sagittal plane is described.
- the body position and posture on the lateral plane orthogonal to the force sagittal plane described above may be corrected.
- the processing from S200 to S218 in FIG. 11, the processing from S300 to S318 in FIG. 12, and the processing from S600 to S618 in FIG. 24 may be extended to three dimensions.
- the correction processing of the body position / posture on the sagittal plane and the correction processing of the body position / posture on the lateral plane are performed independently by the same algorithm as in FIGS. 11, 12, and 24. You can.
- the correction processing of the body position / posture including the vertical component of the body position the correction processing of the body position / posture on the sagittal plane and the correction processing of the body position / posture on the lateral plane
- the vertical component of the body position is corrected by correction on either the sagittal plane or the lateral plane, and the correction on the other plane is performed.
- the vertical component of the body position is removed, and the body position and orientation on the other plane should be corrected! ,.
- the body position / posture in the horizontal plane may be corrected by matching.
- the body position / posture may be corrected for the V, one or two of the sagittal plane, the lateral plane, and the horizontal plane.
- the initial candidates (Pb21_s, ⁇ b21_s) of the first temporary correction body position / posture are set at the time of the previous control cycle.
- the first candidate (Pb21_s, ⁇ b21_s) may be the same as the simplified model gait's body position / posture, for example, although the first temporary corrected body position / posture is determined.
- the initial candidate Pb22-s of the second temporary correction body position is determined using the second temporary correction body position obtained at the time of the previous control cycle.
- the initial candidate Pb22_s may be the same as the upper body position of the simple dani model gait.
- the first temporary correction body position and orientation satisfying the conditions 1 and 2 or the second temporary correction To search for a body position in a short time it is desirable to determine the initial candidates (Pb2_s, ⁇ b2_s) as described in the first to third embodiments.
- the first provisional corrected body position / posture satisfying the above conditions 1 and 2 is searched for, but for example,
- the arrangement of the elements of the two displacement dimension correction model was determined in accordance with the simplified model gait according to the geometric constraint (2).
- the correction amount from the body position / posture of the simplified model gait to the first temporary corrected body position / posture is determined using the function formula or map created in advance, and the correction amount is used to calculate the simplified model gait. By correcting the body position and orientation, it is also possible to determine the first temporary corrected body position and orientation.
- the second temporary The corrected body position was searched for, but the arrangement of the elements of the second displacement dimension correction model (the position of each mass point and the posture of each link with inertia) was determined by the geometric constraint condition (2 ),
- the difference between the arrangement of the simple gait model corresponding to the gait and the arrangement of the elements of the first displacement dimension correction model (the difference in the position of each mass point between the two models and the inertia From the difference in the posture angle of each link), the amount of correction from the upper body position of the simplified model gait to the second temporary correction body position is determined using a function formula or map created in a rough manner.
- the body posture of the simplified model gait is obtained by multiplying the first temporary corrected body posture by the weight wl and the body posture of the simplified model gait.
- the second displacement is calculated from the instantaneous value of the gait replaced by the sum of the gait multiplied by the weight w2 (hereinafter referred to as the replaced gait !!) according to the geometric constraint condition (2).
- the overall center of gravity shift Gc_err between models and the angular movement amount between models are determined. It is not determined whether or not the product deviation Lc_err is within the allowable range (the processing of S210 in FIG. 11), and the search is performed when the number of searches (the number of updates of the candidate (Pb21_s, ⁇ b21_s)) reaches a predetermined number. Upon completion, the candidate at that time (Pb21_s, 0b21_s) may be determined as the first temporary correction body position / posture.
- the search is completed and the candidate at that time (Pb21_s, ⁇ b21_s) may be determined as the first temporary corrected upper body posture.
- the inter-model angular momentum product deviation Lc_err is an allowable range. If the number of searches (the number of updates of the candidate (Pb22_s, ⁇ b22_s)) reaches a predetermined number, the search is not performed (S310 in FIG. 12 or S610 in FIG. 24). Upon completion, the candidate at that time (Pb22_s, 0b22_s) may be determined as the second temporary correction body position / posture.
- the search is completed and the candidate at that time (Pb22_s, ⁇ b22_s) is second temporarily corrected. Please decide it as the body position and posture.
- Equation 08 when calculating the angular momentum product shift Lc_err between the models, for example, the equation 08 is used instead of the equation 07 as described above. You may.
- the terms following ⁇ on the right side of Equation 08 are the line segment connecting the points Ai and Q of the first displacement dimension correction model, and the points Ai and Q of the second displacement dimension correction model. It is a function that changes almost monotonically with the angle (Pil_Q_Pi2) between the line and the line connecting. Therefore, in the first to third embodiments, the expression 08 is used to calculate the angular momentum product shift Lc_err between models, whereby the sixth embodiment of the present invention is constructed. .
- the first and second displacement dimension correcting models are used.
- the thigh mass point A2 of the second displacement dimension correction model with respect to the line connecting the center point of the ankle joint and the center point of the hip joint of each leg 2 A3 (positional deviation in a plane substantially perpendicular to the line segment) or approximately the positional deviation of the center of the knee joint with respect to the line segment.
- the positional deviations (P22-P21), (P32-P31) related to the thigh mass points A2, A3 in the above equations 06 and 07 are obtained.
- the distance between the above line segment and the thigh mass points A2, A3 or the center of each knee joint (hereinafter referred to as the pseudo misalignment distance of the thigh mass points A2, A3) may be used.
- the pseudo displacement of the thigh mass points A2, A3 has a close relationship with the bending angle of the knee joint of each leg 2, and the pseudo displacement of the thigh masses A2, A3 is calculated as the knee displacement. May be determined. More specifically, as shown in FIG.
- each thigh link 24 (the distance between the center points of the hip joint and the knee joint at both ends of the thigh link 24) is L
- the length of the knee joint is The bending angle (the inclination angle of the straight line passing through the center of the knee joint and the center of the ankle joint) with respect to the bending center (the straight line passing through the center of the thigh link (the center passing through the center of the hip joint and the center of the knee joint)) was defined as ⁇ .
- the pseudo displacement distance of each thigh mass point A2, A3 is substantially equal to L * sin (0Z2), where the length L is the same for both thigh links 24, 24.
- L * determined according to the bending angle ⁇ of the knee joint of each leg 2
- sin ( ⁇ Z2) the knee joint of each leg 2 can be used.
- Lower angle can uniquely determined by the geometric model of the robot 1 (link model).
- each of the legs 2 in the first and second displacement dimension correction models has two mass points. It is also possible to construct a displacement dimension correction model that has a mass point in each of the vicinity, lower leg link, and thigh link (thus, each leg 2 has three mass points). In this case, as in the second or third embodiment, when constraining the position of the mass point of the first displacement dimension correction model, two mass points other than each foot mass point are used, for example, the center of the ankle joint and the hip joint. What is necessary is just to set to two points determined by the predetermined internal division ratio on the line segment connecting the center. Also the lower leg link and Z or upper body A rigid body (link) with inertia equivalent to a link may be calorie as an element of both displacement dimension correction models.
- the robot 1 when the operation mode of the robot 1 is the normal mode (running mode and low-friction floor walking mode), the robot 1 is stopped and In order to generate a desired gait that exercises such that the arms 5 and 5 protrude forward together, the first and second displacement dimension correcting models correspond to the respective arms 5. You may want to give the points a mass point and inertia.
- the thigh mass point is calculated by the first and second displacement dimension correcting models in the second and third embodiments.
- a mass point corresponding to the elbow joint or the vicinity thereof may be provided. More specifically, for example, as shown in FIG. 27, in each of the first and second displacement dimension correcting models, in addition to the upper body mass Bl, the thigh mass B2, B3, and the foot mass B4, B5, each arm body
- the elbow mass points B8 and B9 corresponding to the vicinity of the elbow joint 5 and the corresponding hand mass points B6 and B7 corresponding to the vicinity of the distal end of each arm 5 respectively are included in the first displacement dimension correction model.
- the elbow masses B8 and B9 are constrained to a point defined by a predetermined internal division ratio on a line connecting the center of the shoulder joint and the center of the wrist joint of the arm 5. Then, as in the second or third embodiment, the difference between the models including the position difference of the elbow joints B8 and B9 between the first displacement dimension correction model and the second displacement dimension correction model is included.
- the first provisionally corrected body position and posture are calculated so that the total body weight difference Gc_err and the model angular momentum product difference err approach 0 (to satisfy the above conditions 1 and 2), and the body posture is simplified.
- the second temporary correction body position / posture is determined so that the angular momentum product shift err between the models approaches 0 (the condition 2).
- the arm posture of the first displacement dimension correction model is changed to the upright posture of the robot 1 in the same manner as in the first embodiment when the posture of each leg 2 is restricted.
- the arm may be constrained to the arm posture (the posture extended in the vertical direction) in the force state.
- the geometric constraint condition (1) corresponding to the first geometric constraint condition in the present invention and the second geometric constraint condition Since the geometric constraint (2) corresponding to the geometric constraint is set as described above, those geometric constraints (1) and (2) are the ninth aspect of the present invention. It is set as in the invention.
- the geometric constraint (3) corresponding to the first geometric constraint in the present invention and the geometric constraint (3) corresponding to the second geometric constraint in the present invention are used. Since the geometric constraints (2) are set as described above, the geometric constraints (3) and (2) are set as in the eighth invention of the present invention. ing.
- the sum of the masses of all the elements of the first displacement dimension correction model matches the total mass of the robot 1, and the first displacement order relative to the instantaneous target motion of the robot 1 is obtained.
- the overall center-of-gravity position G1 of the original correction model is made to match or almost coincides with the total center-of-gravity position Gs of the simplified model for the instant target movement.
- the sum of the masses of all the elements of the second displacement dimension correction model also matches the total mass of the robot 1, and the overall center of gravity G2 of the second displacement dimension correction model for the instantaneous target movement of the robot 1 is It is made to substantially coincide with the true overall center of gravity Gf of the actual robot 1 for the instantaneous target movement.
- the difference (G1-G2) between G1 and G2 is the difference (Gs ⁇ Gf) between the total weight center position Gs of the simple model and the true overall gravity center position Gf of the robot 1. ), That is, almost the same as the error of the position of the center of gravity of the simplified model. Therefore, the first to third embodiments are characterized in that the geometric constraint condition (1) or (3) as the first geometric constraint condition in the present invention and the second geometric constraint condition as the second geometric constraint condition in the present invention.
- the geometric constraint condition (2) is set as in the ninth invention.
- a mass point such as a mass point near the foot (foot mass point) that is located at the same position in both displacement dimension correction models is regarded as a mass point.
- V may be excluded from both displacement dimension correction models.
- the present invention provides a method of calculating an instantaneous desired gait created using a dynamic model.
- the motion is appropriately corrected by geometrical operations that do not include differential or integral equations, and the dynamic accuracy of the instantaneous target gait including the corrected motion is improved.
- This is useful in that it can provide a gait generating device for a moving-port bot that can minimize fluctuations in the posture of a predetermined part such as a moving part.
- FIG. 1 is a view schematically showing the overall configuration of a mobile robot (bipedal walking robot) to which an embodiment of the present invention is applied.
- FIG. 2 is a side view showing a configuration of a foot portion of each leg of the robot of FIG. 1.
- FIG. 3 is a block diagram showing a configuration of a control unit provided in the robot of FIG. 1.
- FIG. 4 is a block diagram showing a functional configuration of the control unit in FIG. 3.
- FIG. 5 is a block diagram showing functions of the gait generator shown in FIG. 4.
- FIG. 6 is a diagram showing the structure of a simple model (dynamic model) in the first embodiment.
- FIG. 7 (a)-(c) is a diagram showing the relationship between the first displacement dimension correcting model and the simplified model in the first embodiment.
- FIG. 8 is a diagram showing a structure of a second displacement dimension correcting model in the first embodiment.
- FIG. 9 is a flowchart showing a main routine process of the gait generator in the first embodiment.
- FIG. 10 is a flowchart showing processing of a displacement dimension gait correction subroutine in the flowchart of FIG. 9 in the first embodiment.
- FIG. 11 is a flowchart showing a subroutine process of S100 in FIG. 10 in the first embodiment.
- FIG. 12 is a flowchart showing a subroutine process of S102 in FIG. 10 in the first embodiment.
- FIG. 13 is a view for explaining calculation of an angular momentum product in the first embodiment.
- FIG. 14 is a diagram showing an example of the arrangement of elements of the first and second displacement dimension correcting models in the first embodiment.
- FIG. 15 is a diagram showing an example of the arrangement of elements of first and second displacement dimension correcting models according to the first embodiment.
- FIG. 16 is a diagram showing an example of the arrangement of elements of first and second displacement dimension correcting models according to the first embodiment.
- FIG. 17 is a diagram showing an example of a full model used in full model correction.
- FIG. 18 is a diagram showing the structure of a simple model (dynamic model) in the second embodiment.
- FIG. 19 A diagram showing the structure of a first displacement dimension correcting model in a second embodiment.
- FIG. 20 is a diagram showing an example of the arrangement of elements of the first and second displacement dimension correcting models in the second embodiment.
- FIG. 21 is a diagram showing an example of the arrangement of elements of the first and second displacement dimension correcting models in the second embodiment.
- FIG. 22 is a diagram showing an example of the arrangement of elements of the first and second displacement dimension correcting models in the second embodiment.
- FIG. 24 is a flowchart showing the subroutine processing of S506 in FIG. 23.
- FIG. 25 A diagram showing an example of the arrangement of elements of the first and second displacement dimension correcting models in the third embodiment.
- FIG. 26 (a) and (b) are graphs showing examples in which a frequency characteristic is given to the weight wl.
- Fig. 27 is a diagram showing another example of the arrangement of the elements of the first and second displacement dimension correcting models.
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Robotics (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Transportation (AREA)
- Manipulator (AREA)
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2005517747A JP4800037B2 (ja) | 2004-02-06 | 2005-02-04 | 移動ロボットの歩容生成装置 |
US10/597,620 US7715944B2 (en) | 2004-02-06 | 2005-02-04 | Gait generating device of mobile robot |
EP05709758A EP1721711B1 (en) | 2004-02-06 | 2005-02-04 | Gait generator of mobile robot |
KR1020067011906A KR101131773B1 (ko) | 2004-02-06 | 2005-02-04 | 이동 로봇의 보용생성장치 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2004-031271 | 2004-02-06 | ||
JP2004031271 | 2004-02-06 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2005075156A1 true WO2005075156A1 (ja) | 2005-08-18 |
Family
ID=34836042
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2005/001693 WO2005075156A1 (ja) | 2004-02-06 | 2005-02-04 | 移動ロボットの歩容生成装置 |
Country Status (5)
Country | Link |
---|---|
US (1) | US7715944B2 (ja) |
EP (1) | EP1721711B1 (ja) |
JP (1) | JP4800037B2 (ja) |
KR (1) | KR101131773B1 (ja) |
WO (1) | WO2005075156A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2017529246A (ja) * | 2014-08-25 | 2017-10-05 | グーグル インコーポレイテッド | 自然なピッチとロール |
Families Citing this family (26)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101121020B1 (ko) * | 2004-01-13 | 2012-03-15 | 혼다 기켄 고교 가부시키가이샤 | 이동 로봇의 보용생성 장치 |
JP2008073830A (ja) * | 2006-09-25 | 2008-04-03 | Fanuc Ltd | ロボット制御装置 |
US8138707B2 (en) * | 2007-10-23 | 2012-03-20 | Honda Motor Co., Ltd. | Bipedal walking robot |
KR20110017500A (ko) * | 2009-08-14 | 2011-02-22 | 삼성전자주식회사 | 다리식 이동 로봇의 제어장치 및 그 제어방법 |
KR101667032B1 (ko) * | 2009-10-30 | 2016-10-17 | 삼성전자 주식회사 | 로봇의 경로 계획 장치 및 그 방법 |
KR101760883B1 (ko) * | 2010-09-09 | 2017-08-04 | 삼성전자주식회사 | 로봇 및 그 제어방법 |
KR20120071599A (ko) * | 2010-12-23 | 2012-07-03 | 삼성전자주식회사 | 보행 로봇 및 그 제어 방법 |
JP5948932B2 (ja) * | 2012-02-16 | 2016-07-06 | セイコーエプソン株式会社 | ロボット制御装置、ロボット制御方法およびロボット制御プログラムならびにロボットシステム |
US9618937B1 (en) | 2014-08-25 | 2017-04-11 | Google Inc. | Slip detection using robotic limbs |
US9387588B1 (en) | 2014-08-25 | 2016-07-12 | Google Inc. | Handling gait disturbances with asynchronous timing |
US10081098B1 (en) | 2014-08-25 | 2018-09-25 | Boston Dynamics, Inc. | Generalized coordinate surrogates for integrated estimation and control |
US9623556B1 (en) * | 2014-09-03 | 2017-04-18 | X Development Llc | Robotic sole joint |
JP6228097B2 (ja) * | 2014-10-06 | 2017-11-08 | 本田技研工業株式会社 | 移動ロボット |
JP5927270B2 (ja) * | 2014-11-06 | 2016-06-01 | ファナック株式会社 | ロボットシミュレーション装置 |
US9446518B1 (en) * | 2014-11-11 | 2016-09-20 | Google Inc. | Leg collision avoidance in a robotic device |
US9440353B1 (en) | 2014-12-29 | 2016-09-13 | Google Inc. | Offline determination of robot behavior |
US9499218B1 (en) | 2014-12-30 | 2016-11-22 | Google Inc. | Mechanically-timed footsteps for a robotic device |
US9594377B1 (en) * | 2015-05-12 | 2017-03-14 | Google Inc. | Auto-height swing adjustment |
KR101655298B1 (ko) | 2015-09-14 | 2016-09-07 | 홍성훈 | 연통이 구비된 하향식 그릴장치 |
US9586316B1 (en) | 2015-09-15 | 2017-03-07 | Google Inc. | Determination of robotic step path |
US9778132B1 (en) * | 2015-12-16 | 2017-10-03 | X Development Llc | Methods and systems for force sensor calibration |
US9789919B1 (en) | 2016-03-22 | 2017-10-17 | Google Inc. | Mitigating sensor noise in legged robots |
US9987745B1 (en) | 2016-04-01 | 2018-06-05 | Boston Dynamics, Inc. | Execution of robotic tasks |
US20180088234A1 (en) * | 2016-09-27 | 2018-03-29 | Carnegie Mellon University | Robust Localization and Localizability Prediction Using a Rotating Laser Scanner |
CN110053039B (zh) * | 2018-01-17 | 2021-10-29 | 深圳市优必选科技有限公司 | 一种机器人行走中重力补偿的方法、装置及机器人 |
US10946518B2 (en) * | 2018-07-24 | 2021-03-16 | Invia Robotics, Inc. | Spatiotemporal controller for controlling robot operation |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH05237776A (ja) * | 1992-02-26 | 1993-09-17 | Honda Motor Co Ltd | ロボットの関節駆動制御装置 |
JPH05318339A (ja) * | 1992-05-22 | 1993-12-03 | Honda Motor Co Ltd | ロボットのリンクなどの軌道生成方法及び装置 |
JPH0994785A (ja) * | 1995-09-29 | 1997-04-08 | Honda Motor Co Ltd | 脚式歩行ロボットの歩容生成方法 |
JP2002086373A (ja) * | 2000-09-11 | 2002-03-26 | National Institute Of Advanced Industrial & Technology | 脚式ロボットのリアルタイム最適制御方法 |
Family Cites Families (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3148827B2 (ja) | 1992-04-30 | 2001-03-26 | 本田技研工業株式会社 | 脚式移動ロボットの歩行制御装置 |
JP3233450B2 (ja) | 1992-05-22 | 2001-11-26 | 本田技研工業株式会社 | 指定時刻到達関数発生器 |
US5872893A (en) * | 1996-07-25 | 1999-02-16 | Honda Giken Kogyo Kabushiki Kaisha | Gait generation system of legged mobile robot |
JP3663034B2 (ja) | 1996-07-25 | 2005-06-22 | 本田技研工業株式会社 | 脚式移動ロボットの歩容生成装置 |
JP3629133B2 (ja) | 1997-01-31 | 2005-03-16 | 本田技研工業株式会社 | 脚式移動ロボットの制御装置 |
JP3443077B2 (ja) * | 1999-09-20 | 2003-09-02 | ソニー株式会社 | ロボットの運動パターン生成装置及び運動パターン生成方法、並びにロボット |
KR100661333B1 (ko) * | 2000-11-17 | 2006-12-27 | 혼다 기켄 고교 가부시키가이샤 | 레그식 이동 로봇의 보용 생성장치 |
JP3726032B2 (ja) * | 2001-04-27 | 2005-12-14 | 本田技研工業株式会社 | 脚式移動ロボットの目標運動生成装置 |
KR100946709B1 (ko) * | 2001-12-28 | 2010-03-12 | 혼다 기켄 고교 가부시키가이샤 | 다리식 이동로봇의 보용 생성 장치 |
US7496428B2 (en) * | 2001-12-28 | 2009-02-24 | Honda Giken Kogyo Kabushiki Kaisha | Gait producing device for leg type movable robot |
EP1475198B1 (en) | 2002-01-18 | 2007-05-23 | Honda Giken Kogyo Kabushiki Kaisha | Controller of legged mobile robot |
KR101121020B1 (ko) | 2004-01-13 | 2012-03-15 | 혼다 기켄 고교 가부시키가이샤 | 이동 로봇의 보용생성 장치 |
-
2005
- 2005-02-04 KR KR1020067011906A patent/KR101131773B1/ko active IP Right Grant
- 2005-02-04 EP EP05709758A patent/EP1721711B1/en not_active Expired - Fee Related
- 2005-02-04 JP JP2005517747A patent/JP4800037B2/ja not_active Expired - Fee Related
- 2005-02-04 WO PCT/JP2005/001693 patent/WO2005075156A1/ja active Application Filing
- 2005-02-04 US US10/597,620 patent/US7715944B2/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH05237776A (ja) * | 1992-02-26 | 1993-09-17 | Honda Motor Co Ltd | ロボットの関節駆動制御装置 |
JPH05318339A (ja) * | 1992-05-22 | 1993-12-03 | Honda Motor Co Ltd | ロボットのリンクなどの軌道生成方法及び装置 |
JPH0994785A (ja) * | 1995-09-29 | 1997-04-08 | Honda Motor Co Ltd | 脚式歩行ロボットの歩容生成方法 |
JP2002086373A (ja) * | 2000-09-11 | 2002-03-26 | National Institute Of Advanced Industrial & Technology | 脚式ロボットのリアルタイム最適制御方法 |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2017529246A (ja) * | 2014-08-25 | 2017-10-05 | グーグル インコーポレイテッド | 自然なピッチとロール |
Also Published As
Publication number | Publication date |
---|---|
JP4800037B2 (ja) | 2011-10-26 |
EP1721711A1 (en) | 2006-11-15 |
EP1721711A4 (en) | 2009-12-02 |
KR20060126655A (ko) | 2006-12-08 |
EP1721711B1 (en) | 2011-12-07 |
JPWO2005075156A1 (ja) | 2007-10-11 |
US20090171503A1 (en) | 2009-07-02 |
KR101131773B1 (ko) | 2012-04-05 |
US7715944B2 (en) | 2010-05-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP4800037B2 (ja) | 移動ロボットの歩容生成装置 | |
JP5221688B2 (ja) | 脚式移動ロボットの歩容生成装置 | |
JP5219956B2 (ja) | 移動体の制御装置 | |
JP3672101B2 (ja) | 脚式移動ロボットの歩容生成装置 | |
JP4805817B2 (ja) | 移動ロボットの歩容生成装置 | |
JPWO2005000536A1 (ja) | 脚式移動ロボットの制御装置 | |
JP4800038B2 (ja) | 移動ロボットの歩容生成装置 | |
JP5232120B2 (ja) | 移動体の制御装置 | |
JP5404543B2 (ja) | 脚式移動ロボットの制御装置 | |
JP5404544B2 (ja) | 脚式移動ロボットの制御装置 | |
JP2011224757A (ja) | 2足移動ロボットの制御装置及び歩容生成装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NA NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): BW GH GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LT LU MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2005517747 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1020067011906 Country of ref document: KR |
|
WWE | Wipo information: entry into national phase |
Ref document number: 10597620 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWW | Wipo information: withdrawn in national office |
Country of ref document: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2005709758 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 2005709758 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 1020067011906 Country of ref document: KR |