CN105930588A - Physical engine-based robot action generation method - Google Patents
Physical engine-based robot action generation method Download PDFInfo
- Publication number
- CN105930588A CN105930588A CN201610251372.0A CN201610251372A CN105930588A CN 105930588 A CN105930588 A CN 105930588A CN 201610251372 A CN201610251372 A CN 201610251372A CN 105930588 A CN105930588 A CN 105930588A
- Authority
- CN
- China
- Prior art keywords
- robot
- physical engine
- robot model
- frame
- sufficient
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/36—Circuit design at the analogue level
- G06F30/367—Design verification, e.g. using simulation, simulation program with integrated circuit emphasis [SPICE], direct methods or relaxation methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
Abstract
The invention discloses a physical engine-based robot action generation method. The method comprises the following steps of obtaining a robot model by 3D modeling and constructing a simulated ground environment by utilizing a physical engine; setting a weight, gravitational acceleration and a plane friction coefficient of the robot model, and putting the robot model in the simulated ground environment; and executing corresponding actions by the robot model under the gravity, movement inertia or frictional force according to action signals generated by a simulation platform. According to the method, a real movement physical environment of a robot is efficiently simulated in a computer by adopting the simulation platform, the physical engine and the 3D modeling technology, a movement effect that can be achieved only by a real machine test originally is achieved, the robot action generation and test efficiency is improved, and large-scale parallel simulation computing can be carried out without accumulative errors.
Description
Technical field
The present invention relates to robot field, be specifically related to robot motion generation side based on physical engine
Method.
Background technology
The existing debugging to robot motion is mainly taked manually or Program Generating set of actions, then
Set of actions is burnt to robot interior and carries out actual machine test and just can draw the actual running results, tool
The operation of body needs the most man-to-man authentic testing that carries out robot motion, but long-time big
After measuring examination, joint of robot and structural member may need because of heavy wear to adjust or change,
Otherwise error can accumulate and affect test result, and in order to avoid the accumulation of the error shadow to test result
Ring, need to adjust in time or change joint and the structural member of robot.
In sum, the existing debugging technique to robot motion has the disadvantage in that (1) efficiency relatively
Low;(2) need adjust in time or change machine on the impact of test result in order to avoid the accumulation of error
The joint of people and structural member.
In view of this, it is badly in need of designing a kind of new method that robot motion is debugged, it is achieved efficiently
And robot motion is debugged without cumulative error.
Summary of the invention
The technical problem to be solved is to provide a kind of new side debugging robot motion
Method, it is achieved efficiently and without the problem that robot motion is debugged of cumulative error.
In order to solve above-mentioned technical problem, the technical solution adopted in the present invention is to provide a kind of based on thing
The robot motion of reason engine generates method, comprises the following steps:
Utilize 3D modeling to obtain robot model, utilize physical engine constructing analog ground environment;
The weight of robot model, acceleration of gravity and plane coefficient of friction are set, and place it in
In simulation ground environment;
The actuating signal generated by analog platform is controlled robot model and at gravity, motional inertia or rubs
Corresponding action is performed under wiping power.
In technique scheme, in same described simulation ground environment, it is placed with multiple described machine
Device human model.
In technique scheme, described actuating signal includes but not limited to: each joint of each frame rotates
Angle and shift position, each sufficient end of each frame are each relative to the foot coordinate position of coordinate system, each frame
Foot end relative to the coordinate position of fuselage centre coordinate system, each sufficient end of each frame relative to virtual field
The coordinate position of ground level coordinate system, each sufficient independent periods of coordinate position list within the beginning and ending time with
And end rotation signal.
In technique scheme, described physical engine includes but not limited to Havok and PhysX.
In technique scheme, described simulation ground environment includes landform and ground object.
In technique scheme, described robot model includes but not limited to: Six-foot walking robot,
Mechanical arm, anthropomorphic robot and multi-foot robot.
The present invention, uses analog platform, physical engine and 3D modeling technique to achieve in computer-internal
The real motion physical environment of efficient simulation robot, it is achieved that the most only could try by prototype test
The movement effects tested out, improves robot motion and generates and the efficiency of test, can large-scale parallel mould
Intend computing, and without cumulative error.
Accompanying drawing explanation
The robot motion based on physical engine that Fig. 1 provides for the embodiment of the present invention generates method flow
Figure.
Detailed description of the invention
Below in conjunction with specification drawings and specific embodiments, the present invention is described in detail.
Embodiments provide a kind of robot motion based on physical engine and generate method, such as Fig. 1
Shown in, comprise the following steps:
Step 101, utilize 3D modeling obtain robot model, utilize physical engine (such as Havok
And PhysX) constructing analog ground environment.
Step 102, the weight of robot model, acceleration of gravity and plane coefficient of friction are set, and will
It is placed in simulation ground environment.
Step 103, the actuating signal generated by analog platform control robot model in gravity, motion
Corresponding action is performed under inertia or frictional force.
It is placed with multiple robot model in same simulation ground environment.Above-mentioned robot model wraps
Include but be not limited to: Six-foot walking robot, mechanical arm, anthropomorphic robot and multi-foot robot.
Above-mentioned actuating signal includes but not limited to: each joint rotation angle of each frame and shift position, every
The one each sufficient end of frame relative to foot the coordinate position of coordinate system, each sufficient end of each frame relative in fuselage
The each sufficient end of the coordinate position of heart coordinate system, each frame is relative to the coordinate of virtual field ground level coordinate system
The coordinate position list within the beginning and ending time of position, each sufficient independent periods and end rotation signal.
Above-mentioned each frame refers to each frame in the robot whole action cycle, refers to each sufficient independent periods
Be in units of each foot, the action being independently Myself, do not interfered by other foots, such as: No. 1 foot
From the beginning of 0 second, with 3 seconds as cycle, each cycle is equally divided into 9 frames, do one be lifted up-forward
The mobile do action landing-kick backward ground;No. 2 foots are from the beginning of 0.8 second, with 5 seconds as cycle, often
Cycle is equally divided into 15 frames, do one be lifted up-swing twice-drop back into action in situ.1
The action of number foot and No. 2 foots is mutually incoherent, but robot interior program is moved according to the independence of the two foot
Calculate the attitude of each frame of whole robot and perform, being therefore the most then one from robot
Overlap complete action.
Above-mentioned fuselage centre coordinate system is with the fuselage centre point of robot as zero, former through coordinate
The connecting line putting the photographic head place plane being parallel to fuselage plane and being perpendicular to robot head is Y-axis,
Through zero and to be perpendicular to the axle of Y direction be X-axis, through zero and be perpendicular to fuselage plane
Axle be the coordinate system that Z axis is set up.
Above-mentioned virtual field ground level coordinate system is certain point at the central point with virtual field ground level or edge
For zero, virtual field ground level is X-Y plane, through zero and be perpendicular to X-Y plane
Axle is the coordinate system that Z axis is set up, and virtual field ground level coordinate system is believed according to action for robot
Number find target shift position, calculate mobile route.
Above-mentioned sufficient coordinate system is so that often foot root and fuselage junction point are as zero, through zero
Fuselage tangent line is X-axis, with the connection extended line of zero and fuselage centre point as Y-axis, through coordinate
Initial point is perpendicular to the coordinate system that axle is Z axis of X-axis and Y-axis.
In the present invention, with the position of sufficient end each under standard midstance as reference, calculate each
The translated linear path of foot end, due under standard midstance, foot end and fuselage centre point
Position relatively is fixing, so when fuselage is positioned at optional position, can calculate the seat of each sufficient end
Mark, so also just can calculate the path for translation of each sufficient end by fuselage centre point path for translation.
The corresponding relation being set up foot coordinate system and fuselage centre coordinate system by coordinate transformation algorithm is specifically wrapped
Include following steps:
Rotated by robot obtain fuselage centre coordinate system (x, y) coordinate in foot coordinate system (x ',
Y ') be:
Wherein, θ is line and the fuselage centre point in fuselage centre point and robot head photographic head direction
With foot root at the angle of fuselage junction point line;
Carry out translation by robot after being rotated again to obtain fuselage centre coordinate system (x, y, z) at sufficient seat
Mark system in coordinate (x ", y ", z "), wherein x "=x ';Y "=y '+r;R is fuselage half
Footpath, i.e. fuselage centre point to foot root in the distance of fuselage junction point;Owing to z-axis converts,
So z "=z;
Utilize said method, extrapolate each sufficient end position in moving process by the movement of fuselage centre point
The respective coordinates put.
Above-mentioned simulation ground environment includes landform (plane or uneven) and ground object etc.,
Ground object may belong to the part (irremovable) of ground environment or discrete objects (is subject to
After external force impact removable, ratio is if the pillar that knocked down), set each object in simulation ground environment
With weight, acceleration of gravity and the plane coefficient of friction etc. of robot model, the machine then will set up
Human model is placed in simulation ground environment, and the actuating signal generated by analog platform controls robot
Model performs corresponding action under gravity, motional inertia or frictional force, and robot model draws at physics
Hold up and under effect, follow physical law voluntarily.
It is high that the present invention uses analog platform, physical engine and 3D modeling technique to achieve in computer-internal
The real motion physical environment of effect dummy robot, it is achieved that the most only test aptitude tests by prototype
The movement effects gone out, improves robot motion and generates and the efficiency of test, can large-scale parallel simulation
Computing, and without cumulative error.
The present invention is not limited to above-mentioned preferred forms, and anyone should learn the enlightenment in the present invention
Under the structure change made, every have same or like technical scheme with the present invention, each falls within this
Within the protection domain of invention.
Claims (6)
1. robot motion based on physical engine generates method, it is characterised in that comprise the following steps:
Utilize 3D modeling to obtain robot model, utilize physical engine constructing analog ground environment;
The weight of robot model, acceleration of gravity and plane coefficient of friction are set, and place it in
In simulation ground environment;
The actuating signal generated by analog platform is controlled robot model and at gravity, motional inertia or rubs
Corresponding action is performed under wiping power.
2. the method for claim 1, it is characterised in that at same described simulation ground ring
Border is placed with multiple described robot model.
3. the method for claim 1, it is characterised in that described actuating signal includes but do not limits
In: each joint rotation angle of each frame and shift position, each sufficient end of each frame are relative to sufficient coordinate system
Coordinate position, each sufficient end of each frame is relative to the coordinate position of fuselage centre coordinate system, each frame
Each sufficient end relative to the coordinate position of virtual field ground level coordinate system, each sufficient independent periods when start-stop
Interior coordinate position list and end rotation signal.
4. the method for claim 1, it is characterised in that described physical engine includes but do not limits
In Havok and PhysX.
5. the method for claim 1, it is characterised in that described simulation ground environment includes ground
Shape and ground object.
6. the method for claim 1, it is characterised in that described robot model include but not
It is limited to: Six-foot walking robot, mechanical arm, anthropomorphic robot and multi-foot robot.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610251372.0A CN105930588B (en) | 2016-04-21 | 2016-04-21 | Robot motion generation method based on physical engine |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610251372.0A CN105930588B (en) | 2016-04-21 | 2016-04-21 | Robot motion generation method based on physical engine |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105930588A true CN105930588A (en) | 2016-09-07 |
CN105930588B CN105930588B (en) | 2019-04-26 |
Family
ID=56839678
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610251372.0A Active CN105930588B (en) | 2016-04-21 | 2016-04-21 | Robot motion generation method based on physical engine |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105930588B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107895441A (en) * | 2017-10-26 | 2018-04-10 | 广州市雷军游乐设备有限公司 | Obtain method, apparatus, equipment and the storage medium of grabber model seized condition |
CN111251305A (en) * | 2020-03-13 | 2020-06-09 | 南方科技大学 | Robot force control method, device, system, robot and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101493855A (en) * | 2009-01-16 | 2009-07-29 | 吉林大学 | Real-time simulation system for under-driven double-feet walking robot |
US20100049494A1 (en) * | 2008-08-15 | 2010-02-25 | Edsa Corporation | Method for predicting power usage effectiveness and data center infrastructure efficiency within a real-time monitoring system |
CN103345285A (en) * | 2013-06-27 | 2013-10-09 | 山东大学 | Quadruped robot remote control system and remote control method thereof |
CN104200052A (en) * | 2014-09-22 | 2014-12-10 | 哈尔滨工业大学 | System and method for simulating hydraulic-drive hexapod robot |
-
2016
- 2016-04-21 CN CN201610251372.0A patent/CN105930588B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100049494A1 (en) * | 2008-08-15 | 2010-02-25 | Edsa Corporation | Method for predicting power usage effectiveness and data center infrastructure efficiency within a real-time monitoring system |
CN101493855A (en) * | 2009-01-16 | 2009-07-29 | 吉林大学 | Real-time simulation system for under-driven double-feet walking robot |
CN103345285A (en) * | 2013-06-27 | 2013-10-09 | 山东大学 | Quadruped robot remote control system and remote control method thereof |
CN104200052A (en) * | 2014-09-22 | 2014-12-10 | 哈尔滨工业大学 | System and method for simulating hydraulic-drive hexapod robot |
Non-Patent Citations (3)
Title |
---|
宁晗: "仿人机器人实时建模与仿真平台的设计与开发", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
焦立男 等: "多个地面移动机器人建立构型空间", 《计算机仿真》 * |
王小亮 等: "多机器人协同队形变换算法与实现", 《计算机仿真》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107895441A (en) * | 2017-10-26 | 2018-04-10 | 广州市雷军游乐设备有限公司 | Obtain method, apparatus, equipment and the storage medium of grabber model seized condition |
CN111251305A (en) * | 2020-03-13 | 2020-06-09 | 南方科技大学 | Robot force control method, device, system, robot and storage medium |
CN111251305B (en) * | 2020-03-13 | 2023-02-07 | 南方科技大学 | Robot force control method, device, system, robot and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN105930588B (en) | 2019-04-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107256284B (en) | Real-time interactive multi-gait dynamics modeling method and system for quadruped robot | |
CN108058758B (en) | Robot biped walking simulation evaluation system and evaluation method based on six-degree-of-freedom leg | |
Kortier et al. | Hand pose estimation by fusion of inertial and magnetic sensing aided by a permanent magnet | |
CN106625774B (en) | A kind of space manipulator geometrical parameter calibration method | |
Lin et al. | A leg configuration measurement system for full-body pose estimates in a hexapod robot | |
CN101630146B (en) | Simulation control system for remote tele-operation of lunar rover | |
Mistry et al. | Sit-to-stand task on a humanoid robot from human demonstration | |
JP5502348B2 (en) | Simulation method | |
CN104118580A (en) | Device and method for simulating low gravity | |
CN103832504A (en) | Bionic foot-type robot comprehensive simulation strategy | |
US7457733B2 (en) | Moving a virtual articulated object in a virtual environment while avoiding collisions between the articulated object and the environment | |
CN105930588A (en) | Physical engine-based robot action generation method | |
CN105459116B (en) | A kind of teleoperation of robot device and method based on magnetometer | |
Szumowski et al. | Preview control applied for humanoid robot motion generation | |
JP2937834B2 (en) | 3D motion generator | |
Geng | Torso inclination enables faster walking in a planar biped robot with passive ankles | |
CN107665616B (en) | A kind of nine-degree of freedom motion simulator relative motion equivalent method and system | |
Bruneau et al. | Distributed ground/walking robot interaction | |
CN105938499A (en) | Coordinate system establishment method of 3D biped robot | |
KR101635684B1 (en) | Method for Generating and controlling Locomotion, Locomotion Controller | |
CN108874146B (en) | Moving human body mass center displacement calculation method applied to virtual reality system | |
Ootsubo et al. | Support system for slope shaping based on a teleoperated construction robot | |
JP6149791B2 (en) | Center of gravity estimation device and center of gravity estimation method | |
Guo et al. | Mixing control of animating virtual human for maintenance simulation | |
Mihradi et al. | 3D kinematics of human walking based on segment orientation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |