CN101950176A - Method for performing zero moment point (ZMP) calibration autonomously by robot - Google Patents

Method for performing zero moment point (ZMP) calibration autonomously by robot Download PDF

Info

Publication number
CN101950176A
CN101950176A CN2010102699084A CN201010269908A CN101950176A CN 101950176 A CN101950176 A CN 101950176A CN 2010102699084 A CN2010102699084 A CN 2010102699084A CN 201010269908 A CN201010269908 A CN 201010269908A CN 101950176 A CN101950176 A CN 101950176A
Authority
CN
China
Prior art keywords
zmp
robot
ankle
feet
barycenter
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN2010102699084A
Other languages
Chinese (zh)
Other versions
CN101950176B (en
Inventor
黄强
陈学超
余张国
李敬
张伟民
许威
马淦
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Institute of Technology BIT
Original Assignee
Beijing Institute of Technology BIT
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Institute of Technology BIT filed Critical Beijing Institute of Technology BIT
Priority to CN201010269908.4A priority Critical patent/CN101950176B/en
Publication of CN101950176A publication Critical patent/CN101950176A/en
Application granted granted Critical
Publication of CN101950176B publication Critical patent/CN101950176B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Manipulator (AREA)

Abstract

The invention provides a method for performing zero moment point (ZMP) calibration autonomously by a robot. The robot has at least two legs and feet, wherein the at least two legs can drive the robot to walk; the feet are connected with the legs through ankle joints respectively; and the feet are provided with force sensors. The method comprises the following steps of: calculating the projection of a barycenter of the robot on the ground according to each part and the position thereof of the robot, wherein the robot is supported by a single foot to maintain a certain stable fixed posture and the part above the ankle joint of the supporting foot of the robot is seen as a mass point; adjusting the ankle joint of the supporting foot of the robot so as to make the projection of the barycenter spread over the surface of the supporting foot; and obtaining a table by sampling, wherein in the table, a true ZMP corresponds to the ZMP measured by the force sensors. The robot can perform the ZMP calibration according to the projection of the barycenter of the robot and the movement of the ankle joints, so the method has the advantages of no need of human intervention, high efficiency, a large number of sampling points, high precision and low requirement on a mounting environment.

Description

A kind of robot autonomous method that ZMP demarcates of carrying out
Technical field
The present invention relates to a kind ofly to the ZMP of robot scaling method, particularly a kind of anthropomorphic robot independently carries out the scaling method of foot ZMP.
Background technology
Anthropomorphic robot (hereinafter to be referred as robot) is the same with the people, is to move by the walking realization of two legs.Its both legs structure with the mankind seemingly, more traditional wheeled and caterpillar type robot has better maneuverability, especially only has the occasion of the discontinuous contact point that disperses more to embody superiority at broken terrain, stair and with ground.But the biped robot has essential unsettled characteristics, falls down easily.
In order to make robot ambulation, must robot be controlled in real time according to the current oneself state of current environmental information and robot, the change that overcomes environment is with uncertain, make robot can be in actual environment stabilized walking.
When the control apery robot stabilized walking, need know a lot of parameters of anthropomorphic robot, such as the attitude of health, speed, acceleration, barycenter, ZMP is ordered or the like, and wherein ZMP (Zero MomentPoint) has used in known a variety of control methods.So-called ZMP, the moment that is ground force is that point of zero.ZMP is by being installed in the six-dimension force sensor ergometry and the moment of foot, calculating by the ZMP computing formula more at present.Since the measuring error of force transducer, installation environment, many-sided influences such as precision and error after force transducer installs, need be demarcated it, and purpose is to record real ZMP point with force transducer.
Prior art is normally demarcated some fixing point on the robot instep, by translation, and rotation, mathematical measures such as linearization reach the purpose of demarcating whole instep.Specific implementation is at first to choose the point of fixity at robot instep edge, such as eight more special point of fixity as shown in Figure 3, the coordinate of these eight points is known, apply an acting force that makes progress perpendicular to instep for the instep of robot at these points then, the power that six-dimension force sensor measures and the value of moment have been obtained, by the computing formula of ZMP, can draw the coordinate of these points of actual measurement.Often because the influence of each side factor, the coordinate figure that calculates with formula has certain deviation with actual coordinate figure, at this moment by the coordinate system translation, and rotation, mathematical measures such as linearization are coincide the coordinate figure that calculates and actual coordinate figure as far as possible.
Since the measuring error of force transducer, installation environment, many-sided influences such as precision and error; Or some part timing signal of sole needs to correct, and some part does not need to correct, if whole sole is all corrected or do not corrected, causing putting the ZMP that calibrated sensor measurement goes out by these has certain error with real ZMP.And generally need the demarcation personnel to carry out craft and repeatedly demarcate, artificial application of force direction is difficult to reach the requirement perpendicular to instep, and the limited amount of the point of demarcating, and is difficult to cover whole instep, and very consuming time, automaticity is not high.
Summary of the invention
At above-mentioned technical matters of the prior art, the object of the present invention is to provide a kind of method of utilizing the automatic whole instep ZMP of demarcation of center of gravity of robot self.Use this method, robot can demarcate ZMP automatically, and obtains the controlled ZMP of error.
The present invention realizes by the following technical solutions.
A kind of method of the walking mobile device being carried out the ZMP demarcation, described walking mobile device has can drive at least two shanks that its walking is moved, and the foot that is connected by ankle-joint respectively with described shank, described foot is equipped with force transducer, and described method comprises:
Allow described walking mobile device list pin support and keep certain stable anchor, regard the part more than the feet ankle-joint of walking mobile device as a particle, according to described each parts of walking mobile device and position thereof, calculate its barycenter projection on the ground;
Regulate the ankle-joint of the feet of walking mobile device, make the projection of described barycenter travel through the instep of described feet;
By sampling, obtain a true ZMP and the corresponding form of ZMP that goes out by described force sensor measuring.
Above-mentioned and/or others of the present invention are also carried out the system that ZMP demarcates to robot and are realized by providing a kind of, described robot has can drive at least two shanks that its walking is moved, and the foot that is connected by ankle-joint respectively with described shank, described foot is equipped with force transducer, it is characterized in that described system comprises:
Be used to allow described robot list pin support and keep certain stable anchor, regard the part more than the feet ankle-joint of robot as a particle, according to each parts of described robot and position thereof, calculate the device of its barycenter projection on the ground;
Be used to regulate the ankle-joint of the described feet of robot, make the projection of described barycenter travel through the device of the instep of described feet;
Be used for by sampling, obtain the device of a true ZMP and the corresponding form of ZMP that goes out by described force sensor measuring.
Compared with prior art, the present invention has obtained following technique effect:
(1) need not manually to demarcate point of fixity, robot finishes the demarcation of ZMP automatically, has got rid of artificial factor, and is simultaneously time saving and energy saving, the automaticity height.
(2) installation environment no matter, What gives such as precision are as long as situation determines that the ZMP through technical solution of the present invention obtains can both reflect real situation.
(3) number of Cai Yang point is more, and can artificially control, and precision is higher, and the scope of error is controlled.
Description of drawings
Fig. 1 is anthropomorphic robot model simplification figure during according to demarcation ZMP of the present invention.
Fig. 2 is the moving direction figure according to anthropomorphic robot center of gravity of the present invention.
Fig. 3 is that robot of the prior art instep is demarcated synoptic diagram.
Embodiment
Anthropomorphic robot model simplification figure when Fig. 1 illustrates according to demarcation ZMP of the present invention.Robot comprises simplification particle 1, ankle-joint 2, force transducer 3, the instep 4 of feet ankle-joint with top.Ankle-joint can about and fore-and-aft direction rotation.ZMP records by six-dimension force sensor 3.
Fig. 2 is that example shows the direction that center of gravity of the present invention moves with the anthropomorphic robot right crus of diaphragm, partial parameters, the foundation of coordinate system.The sensor upper surface is d to the sole vertical range, p RThe coordinate of expression ZMP.
Before each component-assembled of anthropomorphic robot, the position that can measure its barycenter after assembling is finished, by calculating, can access the barycenter of the robot of any attitude.Suppose to form m by N parts more than the robot anklebone iBe the quality of i parts, p iBe i the position of parts barycenter in the instep coordinate system, the position c that simplifies particle 1 so can calculate by formula (1).
c = ( Σ i = 1 N m i p i ) / Σ i = 1 N m i - - - ( 1 )
At first supporting leg keeps zero-bit, just fully straight leg state, fixing other joints except hip joint (before and after the hip and two joints that swing) by calculating, can obtain making the robot barycenter to pass through the angle of two hip joints of instep coordinate origin in instep projection coordinate.Make two hip joints forward the angle of calculating to, then robot is placed on the level ground, the initial point that the projection of barycenter at this moment just at the feet areal coordinate is.The above part of ankle-joint this moment can be regarded a particle as, and particle can calculate from formula (1) apart from the height L of ankle-joint.
By regulating ankle-joint 2, make the point 5 at barycenter certain edge of sole in the projection arrival Fig. 2 on ground, begin to demarcate ZMP then.The coordinate of postulated point 5 is (x 5, y 5), the angle of ankle-joint left-right rotation is θ, the angle that rotate front and back is
Figure BSA00000253078700041
Can draw by geometric relationship
By top formula, the angular range of ankle-joint that can calculate four marginal points of corresponding instep is [θ Min, θ Max],
Figure BSA00000253078700043
When the barycenter projection of robot moves on to a little 6 from putting 5, only need to change the angle of left-right rotation, the angle that rotate front and back is constant, when the barycenter projection of robot moves on to a little 7 from putting 6, the angle of rotating before and after only needing to change, the angle of left-right rotation is constant, and therefore by changing the angle of ankle-joint, the barycenter projection of robot can travel through convenient whole instep.
By regulating ankle-joint 2, barycenter projection is on the ground moved along direction shown in Figure 2 with speed slowly, simultaneously by force sensor measuring power and moment, power f=[f x, f y, f z], moment τ=[τ x, τ y, τ z], calculate the value of surveying ZMP then, computing formula is as follows:
p Rx=(-τ y-f xd)/f z(2)
p Ry=(τ x-f yd)/f z(3)
p RxBe the value of ZMP in the x direction, p RyBe the value of ZMP in the y direction.Note the position [c of sampling instant barycenter projection on instep x, c y] with the position [p of the ZMP of force transducer actual measurement Rx, p Ry].Allow projection travel through whole sole, can access a barycenter projection and show one to one with actual measurement ZMP.By the time interval and per two width of sampling between the row of regulating sampling, the sparse degree of regulating this table, thereby the size of adjusting maximum error.This table has been divided into a lattice-array to the sole of robot, and in robot actual motion process, we calculate the position p of ZMP by force transducer 3 c=[p x, p y], the mode by tabling look-up obtains Pc and drops in certain grid again, two of grid is P1=[P1x to angular coordinate, P1y], P2=[P2x, P2y] and the position c1=[c1 of each self-corresponding projection x, c1 y], c2=[c2 x, c2 y], satisfy system of equations (4)
P 1 x ≤ P x ≤ P 2 x P 1 y ≤ P y ≤ P 2 y - - - ( 4 )
Calculate the coordinate P of the ZMP that we need at last by formula (5) m=[P Mx, P My].
P mx = c 1 x + ( P x - P 1 x ) / ( P 2 x - P 1 x ) * ( c 2 x - c 1 x ) P my = c 1 y + ( P y - P 1 y ) / ( P 2 y - P 1 y ) * ( c 2 y - c 1 y ) - - - ( 5 )
From foregoing description as can be seen, the invention provides the method that a kind of robot autonomous ZMP of carrying out demarcates, wherein robot carries out the demarcation of ZMP according to the motion of self barycenter projection and ankle-joint, need not artificial interference, the efficient height, the sampled point number is many, the precision height, installation environment requires low.
Embodiments of the invention also can be configured to the ZMP of other robot that can walking moves or moving device is demarcated.Simultaneously, these robot or moving devices that can walking move also are not limited to have the form of two shanks.
Therefore, it will be appreciated by those skilled in the art that the present invention is not limited in the details of this paper description or illustrative method or structure, application or use.In fact, the modification of any appropriate or alternate embodiment all should comprise within the spirit and scope of the present invention.

Claims (8)

1. one kind is carried out the method that ZMP demarcates to the walking mobile device, described walking mobile device has can drive at least two shanks that its walking is moved, and the foot that is connected by ankle-joint respectively with described shank, described foot is equipped with force transducer, it is characterized in that described method comprises:
Allow described walking mobile device list pin support and keep certain stable anchor, regard the part more than the feet ankle-joint of walking mobile device as a particle, according to described each parts of walking mobile device and position thereof, calculate its barycenter projection on the ground;
Regulate the ankle-joint of the feet of walking mobile device, make the projection of described barycenter travel through the instep of described feet;
By sampling, obtain a true ZMP and the corresponding form of ZMP that goes out by described force sensor measuring.
2. method according to claim 1 is characterized in that described force transducer is used for ergometry and moment.
3. method according to claim 2 is characterized in that, when the ankle-joint of the described feet of regulating the walking mobile device, makes the projection of described barycenter overlap with ZMP.
4. method according to claim 3 is characterized in that, described method also comprises:
In actual applications, at first go out ZMP, drop on certain zone of form by searching ZMP that described form obtains measuring then, calculate ZMP at last through having revised by described force sensor measuring.
5. one kind is carried out the system that ZMP demarcates to robot, described robot has can drive at least two shanks that its walking is moved, and the foot that is connected by ankle-joint respectively with described shank, and described foot is equipped with force transducer, it is characterized in that described system comprises:
Be used to allow described robot list pin support and keep certain stable anchor, regard the part more than the feet ankle-joint of robot as a particle, according to each parts of described robot and position thereof, calculate the device of its barycenter projection on the ground;
Be used to regulate the ankle-joint of the described feet of robot, make the projection of described barycenter travel through the device of the instep of described feet;
Be used for by sampling, obtain the device of a true ZMP and the corresponding form of ZMP that goes out by described force sensor measuring.
6. system according to claim 5 is characterized in that described force transducer is used for ergometry and moment.
7. system according to claim 6 is characterized in that, when the ankle-joint of the described feet of regulating robot, makes the projection of described barycenter overlap with ZMP.
8. system according to claim 7 is characterized in that, described system also comprises:
Be used in actual applications, go out the device of ZMP by described force sensor measuring
Be used for by searching the device that ZMP that described form obtains measuring drops on certain zone of form,
Be used to calculate device through the ZMP that has revised.
CN201010269908.4A 2010-09-02 2010-09-02 Method for performing zero moment point (ZMP) calibration autonomously by robot Active CN101950176B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201010269908.4A CN101950176B (en) 2010-09-02 2010-09-02 Method for performing zero moment point (ZMP) calibration autonomously by robot

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201010269908.4A CN101950176B (en) 2010-09-02 2010-09-02 Method for performing zero moment point (ZMP) calibration autonomously by robot

Publications (2)

Publication Number Publication Date
CN101950176A true CN101950176A (en) 2011-01-19
CN101950176B CN101950176B (en) 2014-12-10

Family

ID=43453692

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201010269908.4A Active CN101950176B (en) 2010-09-02 2010-09-02 Method for performing zero moment point (ZMP) calibration autonomously by robot

Country Status (1)

Country Link
CN (1) CN101950176B (en)

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104298243A (en) * 2014-08-19 2015-01-21 北京理工大学 Humanoid robot uneven ground walking stability control method
CN104570732A (en) * 2014-12-15 2015-04-29 浙江大学 Online generation method for walking mode of humanoid robot
CN104634486A (en) * 2014-12-26 2015-05-20 清华大学 Walking robot foot strength measuring device
CN105511465A (en) * 2015-12-02 2016-04-20 歌尔声学股份有限公司 Gait control method and device for dual-foot robot
CN104238360B (en) * 2014-09-03 2016-08-24 浙江大学 A kind of acquisition methods of the Zc parameter of anthropomorphic robot desk-little vehicle model
US9778132B1 (en) 2015-12-16 2017-10-03 X Development Llc Methods and systems for force sensor calibration
CN108186017A (en) * 2017-11-30 2018-06-22 北京理工大学 A kind of detecting system and method for being used to determine pose in capsule endoscope body
CN108237531A (en) * 2016-12-26 2018-07-03 电子科技大学中山学院 A kind of humanoid robot gait's Learning Control Method
CN108345211A (en) * 2017-01-23 2018-07-31 深圳市祈飞科技有限公司 Biped anthropomorphic robot and its non-linear gait planning method and control method
CN110083075A (en) * 2019-05-15 2019-08-02 桂林电子科技大学 A kind of the stability margin evaluation method and control method of pulling type bicycle
CN110244791A (en) * 2019-07-11 2019-09-17 北京理工大学 A kind of biped robot's foot power and moment follow-up control method
CN110802593A (en) * 2019-11-07 2020-02-18 北京理工大学 Lower limb joint zero calibration method of humanoid robot
CN114115295A (en) * 2021-12-28 2022-03-01 深圳市优必选科技股份有限公司 Robot motion planning method and device, robot control equipment and storage medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050267630A1 (en) * 2002-09-26 2005-12-01 Shuuji Kajita Walking gait producing device for walking robot
WO2007046568A1 (en) * 2005-10-19 2007-04-26 Korea Institute Of Science And Technology A method for controlling the walk of humanoid robot
CN101219540A (en) * 2006-10-17 2008-07-16 双叶电子工业株式会社 Robotic system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050267630A1 (en) * 2002-09-26 2005-12-01 Shuuji Kajita Walking gait producing device for walking robot
WO2007046568A1 (en) * 2005-10-19 2007-04-26 Korea Institute Of Science And Technology A method for controlling the walk of humanoid robot
CN101219540A (en) * 2006-10-17 2008-07-16 双叶电子工业株式会社 Robotic system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
刘莉等: "基于六维力/力矩传感器的拟人机器人实际ZMP检测", 《机器人》 *
龚佳伟等: "具有拟人跨步的助步轮椅重心检测", 《上海大学学报(自然科学版)》 *

Cited By (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104298243A (en) * 2014-08-19 2015-01-21 北京理工大学 Humanoid robot uneven ground walking stability control method
CN104298243B (en) * 2014-08-19 2017-02-15 北京理工大学 Humanoid robot uneven ground walking stability control method
CN104238360B (en) * 2014-09-03 2016-08-24 浙江大学 A kind of acquisition methods of the Zc parameter of anthropomorphic robot desk-little vehicle model
CN104570732A (en) * 2014-12-15 2015-04-29 浙江大学 Online generation method for walking mode of humanoid robot
CN104570732B (en) * 2014-12-15 2017-02-08 浙江大学 Online generation method for walking mode of humanoid robot
CN104634486A (en) * 2014-12-26 2015-05-20 清华大学 Walking robot foot strength measuring device
CN105511465A (en) * 2015-12-02 2016-04-20 歌尔声学股份有限公司 Gait control method and device for dual-foot robot
WO2017092463A1 (en) * 2015-12-02 2017-06-08 歌尔股份有限公司 Method of controlling walking posture of biped robot and device
US10031524B2 (en) 2015-12-02 2018-07-24 Goertek Inc. Method and device for controlling gait of biped robot
US9778132B1 (en) 2015-12-16 2017-10-03 X Development Llc Methods and systems for force sensor calibration
CN108237531A (en) * 2016-12-26 2018-07-03 电子科技大学中山学院 A kind of humanoid robot gait's Learning Control Method
CN108345211A (en) * 2017-01-23 2018-07-31 深圳市祈飞科技有限公司 Biped anthropomorphic robot and its non-linear gait planning method and control method
CN108186017A (en) * 2017-11-30 2018-06-22 北京理工大学 A kind of detecting system and method for being used to determine pose in capsule endoscope body
CN110083075A (en) * 2019-05-15 2019-08-02 桂林电子科技大学 A kind of the stability margin evaluation method and control method of pulling type bicycle
CN110083075B (en) * 2019-05-15 2022-04-08 桂林电子科技大学 Stability margin estimation method and control method for pull-type bicycle
CN110244791A (en) * 2019-07-11 2019-09-17 北京理工大学 A kind of biped robot's foot power and moment follow-up control method
CN110244791B (en) * 2019-07-11 2020-05-15 北京理工大学 Foot force and moment following control method for biped robot
CN110802593A (en) * 2019-11-07 2020-02-18 北京理工大学 Lower limb joint zero calibration method of humanoid robot
CN114115295A (en) * 2021-12-28 2022-03-01 深圳市优必选科技股份有限公司 Robot motion planning method and device, robot control equipment and storage medium
CN114115295B (en) * 2021-12-28 2023-08-18 深圳市优必选科技股份有限公司 Robot motion planning method and device, robot control equipment and storage medium

Also Published As

Publication number Publication date
CN101950176B (en) 2014-12-10

Similar Documents

Publication Publication Date Title
CN101950176B (en) Method for performing zero moment point (ZMP) calibration autonomously by robot
US11027415B1 (en) Generalized coordinate surrogates for integrated estimation and control
US8504208B2 (en) Mobile object controller and floor surface estimator
Hoover et al. Bio-inspired design and dynamic maneuverability of a minimally actuated six-legged robot
EP3186041B1 (en) Natural pitch and roll
Gehring et al. Dynamic trotting on slopes for quadrupedal robots
Fan et al. Data fusion for indoor mobile robot positioning based on tightly coupled INS/UWB
CN105137969B (en) Quadruped robot trot gait body posture control methods based on support line Kinematic Decomposition
CN101943912A (en) Control method for eliminating upper body posture shaking of double-foot humanoid robot
KR101272193B1 (en) Legged mobile robot and control program
KR101140810B1 (en) Gait generator for mobile robot
Yuan et al. 3-D localization of human based on an inertial capture system
CN102670217A (en) Wearable sensor measuring device and method for lower limb joint acting force and moment
Reinstein et al. Dead reckoning in a dynamic quadruped robot: Inertial navigation system aided by a legged odometer
US9778132B1 (en) Methods and systems for force sensor calibration
CN112051741A (en) Dynamic motion generation and control method for biped robot
CN103112517B (en) A kind of method and apparatus regulating quadruped robot body posture
Tavakoli et al. Motion control of an omnidirectional climbing robot based on dead reckoning method
CN106225801A (en) A kind of method of personnel's step-length based on inertia sensing estimation
KR20150092332A (en) Method and system for controlling a flying wing
Kuehn et al. Additional DOFs and sensors for bio-inspired locomotion: Towards active spine, ankle joints, and feet for a quadruped robot
CN202932930U (en) Wearable sensor measurement device for acting forces and moments related to lower limb joints of human body
Sandy et al. Dynamically decoupling base and end-effector motion for mobile manipulation using visual-inertial sensing
CN210757742U (en) Foot state detection device of biped robot
CN105549592A (en) Intelligent spherical robot with fault tolerance control and control method of robot

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant