CN107818318A - A kind of anthropomorphic robot imitates method for evaluating similarity - Google Patents
A kind of anthropomorphic robot imitates method for evaluating similarity Download PDFInfo
- Publication number
- CN107818318A CN107818318A CN201711202768.7A CN201711202768A CN107818318A CN 107818318 A CN107818318 A CN 107818318A CN 201711202768 A CN201711202768 A CN 201711202768A CN 107818318 A CN107818318 A CN 107818318A
- Authority
- CN
- China
- Prior art keywords
- mover
- connecting rod
- joint
- msub
- mrow
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J11/00—Manipulators not otherwise provided for
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/008—Artificial life, i.e. computing arrangements simulating life based on physical entities controlled by simulated intelligence so as to replicate intelligent life forms, e.g. based on robots replicating pets or humans in their appearance or behaviour
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Manipulator (AREA)
Abstract
The invention discloses a kind of anthropomorphic robot to imitate method for evaluating similarity, comprises the following steps:1) three dimensional local information of skeleton node is obtained;2) skeleton vector is built, and human visual joint is established according to skeleton vector sum robot joint structure, forms human skeleton model;3) the connecting rod reference frame based on each connecting rod of human skeleton model is established;4) the connecting rod bone vector of human skeleton model is transformed into its female connecting rod reference frame;5) vectorial coordinate of each connecting rod of calculating robot in its female link rod coordinate system;6) similarity of anthropomorphic robot and human body attitude is calculated according to the vectorial coordinate of bone vectorial coordinate and robot links in its female link rod coordinate system of the human skeleton model connecting rod in its female link rod coordinate system.Connecting rod vector quantization evaluation posture similarity of the present invention using connecting rod in its female link rod coordinate system, have characterize fully, stress the characteristics of local, evaluation is accurate.
Description
Technical field
The present invention relates to human-computer interaction technique field, more particularly to a kind of anthropomorphic robot to imitate method for evaluating similarity.
Background technology
In recent years, robot technology is fast-developing, and is applied to industry, medical treatment, scientific research, education more and more widely
Among the field such as training and family daily life.At the same time, various application environment that becomes increasingly complex is fitted to robot
Answering property and intellectuality propose higher requirement.And robot learning by imitation technology can improve robot learning efficiency, lifting
Robot automtion degree, developer is freed from heavy programing work.Therefore, posture and people are imitated to robot
The similarity of body posture carries out appropriate evaluation, and the optimization for robot imitation system has the function that important.It is current
Research is using the method for subjective assessment, and this kind of method need to provide questionnaire to collect information, and process is relatively complicated, it is difficult to obtains essence
True quantitatively evaluating index.Other research is imitated similarity to robot using method for quantitatively evaluating and evaluated, these
Method is mostly based on joint angles or body key node.And joint angles or body key node can not directly characterize imitation
Or the posture of demonstration, thus it is difficult to the similarity of effective evaluation anthropomorphic robot action imitation.
The content of the invention
In order to solve in the prior art based on joint angles or body key node can not directly characterize robot imitate or
The posture of demonstration, and it is difficult to the deficiency of effective evaluation anthropomorphic robot action imitation similarity, the present invention provides a kind of company of utilization
The anthropomorphic robot that bar vector is characterized to the posture of robot and human body imitates method for evaluating similarity.
The purpose of the present invention can be achieved through the following technical solutions:
A kind of anthropomorphic robot imitates method for evaluating similarity, comprises the following steps:
1st, the three dimensional local information of skeleton node is obtained by depth camera;
2nd, according to the three dimensional local information of above-mentioned skeleton node construct skeleton vector, and according to skeleton to
Amount and robot joint structure establish human visual joint, form human skeleton model;
3rd, the connecting rod reference frame based on each connecting rod of human skeleton model is established;
4th, the connecting rod bone vector of human skeleton model is transferred in its female connecting rod reference frame using transition matrix;
5th, according to each joint angles of robot and corresponding each connecting rod of spin matrix calculating robot in its female link rod coordinate system
In vectorial coordinate;
6th, existed according to bone vectorial coordinate of the human skeleton model connecting rod in its female link rod coordinate system and robot links
Vectorial coordinate in its female link rod coordinate system calculates the similarity of anthropomorphic robot and human body attitude.
Further, the skeleton node refers to the artis that can be rotated in bone and bone end segment
Point.Specifically it may include:Vertebra bottom, vertebra middle part, vertebra top, neck, head, left shoulder, left elbow, left wrist, left hand, left hand
Sharp, left thumb, right shoulder, right elbow, right wrist, the right hand, the right hand be sharp, right thumb, left hip, left knee, left ankle, left foot, right hip, right knee, the right side
Ankle, right crus of diaphragm.Further, the robot joint structure refers to the joint rotation that can be rotated in anthropomorphic robot structure
The number of axle and position characteristics, establish human visual joint and refer in skeleton nodal basis, establish and joint of robot number
The amount virtual joint corresponding with type, these virtual joints form the joint in human skeleton model.Consequently facilitating follow-up appearance
The comparison of state similarity.
Further, human skeleton model is defined as follows:By trunk, head, left upper arm, lower-left arm, left hand, left thigh,
Left leg, left foot, right upper arm, bottom right arm, the right hand, right thigh, right leg, right crus of diaphragm are defined as connecting rod;It will be close to the connecting rod of trunk
It is defined as female connecting rod away from trunk connecting rod;The connecting rod that will be far from trunk is defined as the sub-link close to trunk connecting rod;By neck,
Left shoulder, left elbow, left wrist, left hip, left knee, left ankle, right shoulder, right elbow, right wrist, right hip, right knee, right ankle are defined as main joint;By people
The different frees degree possessed by anthropomorphic robot main joint are defined as the secondary joint of the main joint;It will be close to the main joint or secondary of trunk
Joint is defined as away from the main joint of trunk or the female joint in secondary joint;The main joint or secondary joint that will be far from trunk are defined as leaning on
The nearly main joint of trunk or the sub- joint in secondary joint;Connecting rod in two connecting rods being connected with a joint close to trunk is defined as
Female connecting rod in the joint, the connecting rod away from trunk are defined as the sub-link in the joint;Two joints being connected with a connecting rod
In close to the joint of trunk be defined as the female joint of the connecting rod, the joint away from trunk is defined as the sub- joint of the connecting rod.
Further, it is described to establish human visual joint according to skeleton vector sum robot joint structure, form people
Body skeleton model, is specifically included:Established in human skeleton model and anthropomorphic robot quantity and type identical virtual joint.
According to the joint feature of general anthropomorphic robot, in initial attitude, the secondary joint rotary shaft of virtual joint and the mother in the joint
The bone collinear vector of connecting rod is vertical, if two joints belong to a main joint together, two times joint is mutually perpendicular to.This hair
The secondary joint of bright consideration main joint is the situation of one or two.
Further, the virtual joint type includes:Rolling, pitching, driftage;The initial attitude is defined as under both arms
Vertical upright posture.
Further, the connecting rod reference frame of the foundation based on each connecting rod of human skeleton model, is specifically included:For
Sub- joint is the connecting rod of the main joint containing two joints, and two axles of its reference frame are respectively with two sub- joints first
Rotary shaft during beginning joint angles is conllinear, and another rotary shaft is vertical with the plane that the two axles are formed, and forms right hand right angle and sits
Mark system;In initial attitude, X-axis points to the front of human skeleton model, and Y-axis points to the front-left of human skeleton model, Z axis
Point to the surface of human skeleton model.It is the connecting rod of the main joint containing a time joint for sub- joint, its reference coordinate
One reference axis of system is conllinear with the rotary shaft in sub- joint, and a reference axis and the connecting rod bone collinear vector or vertical are another
Individual rotary shaft is vertical with the plane that the two axles are formed, and forms right hand rectangular coordinate system;In initial attitude, X-axis points to human body
The surface of skeleton model, Y-axis point to the front-left of human skeleton model, and Z axis points to the surface of human skeleton model.Connecting rod
The origin of coordinates of coordinate system is respectively positioned on connecting rod end point, i.e., at the sub- articulation center point of connecting rod.
It is further, described that the connecting rod bone vector of human skeleton model is transferred in its female connecting rod reference frame,
Specifically include:Utilize transition matrix RmConnecting rod bone vector in depth camera coordinate system is transformed into the connecting rod of its female connecting rod
In coordinate system, transition matrix RmFor:
WhereinIt is positive with depth camera coordinate system X-axis, Y-axis, Z axis respectively in depth camera coordinate system
Vector in the same direction,Connecting rod respectively in depth camera coordinate system with target connecting rod mother's connecting rod is with reference to seat
The positive vector in the same direction of mark system X-axis, Y-axis, Z axis;
For k-th of human skeleton model connecting rod for being used to calculate posture similarity, in the connecting rod referential of its female connecting rod
In connecting rod bone vector(column vector) is:
Wherein RmkFor corresponding transition matrix,For the connecting rod bone vector in depth camera coordinate system.
Further, it is described according to each each connecting rod of joint angles calculating robot of robot in female link rod coordinate system to
Coordinate is measured, is specifically included:Make initial attitude connecting rod vectorial coordinate of each connecting rod of robot in its female link rod coordinate system be (0,
0,1)T, according to the anglec of rotation θ in each secondary joint of its female jointmki, utilize corresponding spin matrix R (θmki), obtain current pose
Connecting rod vectorial coordinate
Wherein, n is the secondary joint number of target connecting rod female joint, and i numerical value is smaller to represent this joint in connecting rod-joint
Closer to female connecting rod in chain.
Further, the bone vectorial coordinate and machine according to human skeleton model connecting rod in its female link rod coordinate system
Vectorial coordinate of the device people connecting rod in its female link rod coordinate system calculates anthropomorphic robot and the similarity of humanoid posture, specific bag
Include:According to conversion coordinate system after connecting rod vector, using human skeleton model connecting rod bone vector sum corresponding machine people connecting rod to
The cosine value of the angle of amount calculates the coefficient correlation of two connecting rod attitudes, calculates the arithmetic average of each group connecting rod coefficient correlation afterwards
Value, obtain the index of similarity of robot motion imitation.
The present invention has following beneficial effect relative to prior art:
1st, the connecting rod reference frame of each connecting rod of human skeleton model is established, each connecting rod of human body can be accurately determined
Posture.
2nd, using quantitative evaluation method, the precise quantification index of robot pose and human body attitude similarity can be obtained.
3rd, it is more abundant and effective to carry out sign to the posture of robot and human body using connecting rod vector, can be obtained
More accurate posture similarity evaluation.
4th, the connecting rod vector using connecting rod in its female link rod coordinate system, can be with effective evaluation local pose similarity.
Brief description of the drawings
Fig. 1 is that a kind of anthropomorphic robot of the present invention imitates the flow chart of method for evaluating similarity;
Fig. 2 is by the Kinect II skeleton nodenames obtained and numbering schematic diagram;
Fig. 3 is Nao robots and human skeleton model general joint and link rod coordinate system schematic diagram.
Embodiment
Below in conjunction with embodiments of the invention and accompanying drawing, the invention will be further described, but the effect of the present invention is not
It is limited to this:
Embodiment
The invention discloses the evaluation method that a kind of anthropomorphic robot imitates similarity, depth is used as using Kinect II
Camera, anthropomorphic robot is used as using Nao robots.
Anthropomorphic robot imitates the embodiment of method for evaluating similarity, the step flow chart of implementation as shown in figure 1,
Specifically comprise the following steps:
S1, the three dimensional local information by depth camera acquisition skeleton node;
The three-dimensional information of skeleton node, each bone nodename of acquisition are obtained using Kinect II cameras
And numbering is as shown in Figure 2.
S2, skeleton vector is built according to skeleton node, and according to skeleton vector sum joint of robot knot
The vertical human visual joint of structure, forms human skeleton model;
Skeleton vector is established according to the three dimensional local information of skeleton node.It is located at depth camera coordinate system
In, the bone vector that another bone node is pointed to by a bone node turns toWherein a is the bone section at vectorial top
The digital number of point, b are the digital number of the bone node of vector extremities, and bony segment vector is as shown in Figure 3.
According to the articulation structure feature of Nao robots, in human skeleton model structure and Nao joint of robot quantity with
The consistent virtual joint of type, because driving and left and right are gone in the driftage pitching joint of Nao robot trunks bottom by a motor
Symmetrically, therefore in this link do not consider.The original state for defining robot and human skeleton model is the upright of arms down
Posture, now corresponding joint angles be defined as initial joint angles, wherein shoulder pitch angle isRemaining joint angles is
0.In initial attitude the rotary shaft of virtual joint with bone collinear vector or vertical.
By taking left half body partial joint as an example, as shown in figure 3, structure human skeleton model virtual joint, remaining joint is similarly.
Left shoulder pitching joint is the female joint in left shoulder rolling joint, its rotary shaft and vectorCollinearly, left shoulder rolling joint
Rotary shaft position change with the change in left shoulder pitching joint, its rotary shaft and plane 2-5-6 normal vectorAltogether
Line,For:When left shoulder pitching joint is initial angle, the rotation in left shoulder rolling joint
Axle and the normal vector of upper left trunk reference planesCollinearly,For:
Left elbow driftage joint is the female joint in left elbow rolling joint, its rotary shaft and amountCollinearly.Left elbow rolling joint
Rotary shaft position with left elbow go off course joint change and change, its rotary shaft and plane 5-6-7 normal vectorAltogether
Line,For:When left elbow driftage joint is initial angle, the rotation in left elbow rolling joint
The normal vector of axle and plane 2-5-6Collinearly.
Left hip rolling joint is the normal vector of the female joint in left hip pitching joint, its rotary shaft and lower-left trunk reference planesCollinearly,For:The rotary shaft position in left hip pitching joint turns over left hip
Roll joint change and change, its rotary shaft with byWithThe planar process collinear vector of composition.When left hip rolls
When joint is initial angle, rotary shaft and the vector in left hip pitching jointCollinearly.
Left knee pitching joint is left knee uniquely secondary joint, its rotary shaft and plane 17-18-19 normal vector
Collinearly,For:
S3, establish the connecting rod reference frame based on each connecting rod of human skeleton model;
By taking the connecting rod of left half body part as an example, as shown in figure 3, the connecting rod referential of structure human skeleton model connecting rod, remaining
Connecting rod is similarly.
The rotation in the left shoulder rolling joint when X-axis of upper left trunk connecting rod referential and left shoulder pitching joint are initial angle
Axle is conllinear, points to the front of connecting rod;Y-axis is conllinear with the rotary shaft in left shoulder pitching joint, points to the front-left of connecting rod;Z axis hangs down
Directly in X-Y plane, the surface of connecting rod is pointed to.Its origin of coordinates is located at bone node 5.
The rotary shaft in the left elbow rolling joint when X-axis of left big arm link referential and left elbow driftage joint are initial angle
Collinearly, the front of connecting rod is pointed to;Z axis is conllinear with the rotary shaft in left elbow driftage joint, points to the surface of connecting rod;Y-axis is vertical
In X-Z plane, the front-left of connecting rod is pointed to.Its origin of coordinates is located at bone node 6.
The X-axis of lower-left trunk connecting rod referential is conllinear with the rotary shaft in left hip rolling joint, points to the front of connecting rod;Y
The rotary shaft in left hip pitching joint when axle and left hip rolling joint are initial angle is conllinear, points to the front-left of connecting rod;Z axis hangs down
Directly in X-Y plane, the front-left of connecting rod is pointed to.Its origin of coordinates is located at bone node 17.
The Y-axis of left thigh connecting rod referential is conllinear with the rotary shaft in left knee pitching joint, points to the front-left of connecting rod;Y-axis
With left thigh connecting rod bone vectorCollinearly, the surface of connecting rod is pointed to;X-axis is pointing to connecting rod just perpendicular to Y-Z plane
Front.Its origin of coordinates is located at bone node 18.
It is the front of connecting rod when orientation and initial attitude of the front, front-left, surface of connecting rod in connecting rod, positive left
Side, surface are consistent respectively.
S4, the connecting rod bone vector of human skeleton model is transformed into its female connecting rod reference frame;
In order to which the connecting rod bone vector in depth camera coordinate system is transformed into the link rod coordinate system of its female connecting rod, structure
Make transition matrix RmFor:
WhereinRespectively in depth camera coordinate system with depth camera coordinate system X-axis, Y-axis, Z axis just
To vector in the same direction,Connecting rod with target connecting rod mother's connecting rod respectively in depth camera coordinate system refers to
The positive vector in the same direction of coordinate system X-axis, Y-axis, Z axis.
Then for k-th of human skeleton model connecting rod for being used to calculate posture similarity, referred in the connecting rod of its female connecting rod
Connecting rod bone vector in system(column vector) is:
Wherein RmkFor corresponding transition matrix,For the connecting rod bone vector in depth camera coordinate system.
S5, the vectorial coordinate according to each each connecting rod of joint angles calculating robot of robot in its female link rod coordinate system;
Make initial attitude connecting rod of k-th of the connecting rod for being used to calculate posture similarity of robot in its female link rod coordinate system
Vectorial coordinate isAccording to the anglec of rotation θ in each secondary joint of its female jointmki, utilize corresponding spin matrix R
(θmki), obtain the connecting rod vectorial coordinate of current pose
Wherein, n is the secondary joint number of target connecting rod female joint, and i numerical value is smaller to represent this joint in connecting rod-joint
Closer to female connecting rod in chain.
S6, bone vectorial coordinate and robot links seat according to human skeleton model connecting rod in its female link rod coordinate system
Vectorial coordinate in mark system calculates the similarity of anthropomorphic robot and human body attitude.
Further, according to the step S6, the phase relation of k-th of connecting rod of k-th of connecting rod of robot and human skeleton model
NumberFor:
In the present embodiment, trunk, head, left upper arm, lower-left arm, left hand, left thigh, left leg, right upper arm, the right side are considered
10 connecting rods such as underarm, the right hand, right thigh, right leg, stress the anthropomorphic robot and human body attitude similarity of local similarity
Index is:
Described above is the preferred embodiment of the invention, but the protection domain of patent of the present invention is not limited thereto, any
Those familiar with the art in the scope disclosed in patent of the present invention, according to the inventive concept of patent of the present invention or
Person's technical scheme is subject to equivalent substitution or change, belongs to the protection domain of patent of the present invention.
Claims (10)
1. a kind of anthropomorphic robot imitates method for evaluating similarity, it is characterised in that:Methods described comprises the following steps:
1) three dimensional local information of skeleton node is obtained by depth camera;
2) skeleton vector is constructed according to the three dimensional local information of skeleton node, and according to skeleton vector sum machine
Person joint's structure establishes human visual joint, forms human skeleton model;
3) the connecting rod reference frame based on each connecting rod of human skeleton model is established;
4) the connecting rod skeleton model vector of human skeleton model is transformed into its female connecting rod reference frame;
5) vectorial coordinate according to each each connecting rod of joint angles calculating robot of robot in its female link rod coordinate system;
6) it is female at it according to bone vectorial coordinate of the human skeleton model connecting rod in its female link rod coordinate system and robot links
Vectorial coordinate in link rod coordinate system calculates the similarity of anthropomorphic robot and human body attitude.
2. a kind of anthropomorphic robot according to claim 1 imitates method for evaluating similarity, it is characterised in that:The human body
Bone node refers to the artis that can be rotated in bone and bone endpoint node.
3. a kind of anthropomorphic robot according to claim 1 imitates method for evaluating similarity, it is characterised in that:The basis
The three dimensional local information construction skeleton vector of skeleton node, and according to skeleton vector sum robot joint structure
Human visual joint is established, human skeleton model is formed, specifically includes:
Foundation and anthropomorphic robot amount of articulation and type identical virtual joint in human skeleton model, according to humanoid machine
The joint feature of people, in initial attitude, the secondary joint rotary shaft of virtual joint and the bone vector of female connecting rod in the joint are common
Line is vertical, if two joints belong to a main joint together, two times joint is mutually perpendicular to.
4. a kind of anthropomorphic robot according to claim 3 imitates method for evaluating similarity, it is characterised in that:It is described virtual
Joint type includes:Rolling, pitching, driftage;The initial attitude is defined as the upright posture of arms down, corresponding joint
Angle is defined as initial joint angles.
5. a kind of anthropomorphic robot according to claim 1 imitates method for evaluating similarity, it is characterised in that:It is described humanoid
The composition part of robot and humanoid skeleton model, is specifically included:
By the trunk of anthropomorphic robot and humanoid skeleton model, head, left upper arm, lower-left arm, left hand, left thigh, left leg, a left side
Pin, right upper arm, bottom right arm, the right hand, right thigh, right leg, right crus of diaphragm are defined as connecting rod;Will be close to trunk connecting rod be defined as it is remote
Female connecting rod of the connecting rod of trunk;The connecting rod that will be far from trunk is defined as the sub-link of the connecting rod close to trunk;By neck, left shoulder,
Left elbow, left wrist, left hip, left knee, left ankle, right shoulder, right elbow, right wrist, right hip, right knee, right ankle are defined as main joint;By humanoid machine
The different frees degree possessed by people's main joint are defined as the secondary joint of the main joint;The main joint or secondary joint that will be close to trunk are determined
Justice is away from the main joint of trunk or the female joint in secondary joint;It will be far from the main joint of trunk or secondary joint be defined as close to trunk
Main joint or secondary joint sub- joint;Connecting rod in two connecting rods being connected with a joint close to trunk is defined as the pass
Female connecting rod of section, the connecting rod away from trunk are defined as the sub-link in the joint;Leaned in two joints being connected with a connecting rod
The joint of nearly trunk is defined as the female joint of the connecting rod, and the joint away from trunk is defined as the sub- joint of the connecting rod.
6. a kind of anthropomorphic robot according to claim 5 imitates method for evaluating similarity, it is characterised in that:The basis
The distance of connecting rod or joint and trunk judges closer or far from its basis for estimation is:In connecting rod-articulated chain weigh connecting rod or
Joint and the distance of trunk, rather than connecting rod or the distance of joint and trunk are weighed in the total space.
7. a kind of anthropomorphic robot according to claim 1 imitates method for evaluating similarity, it is characterised in that:The foundation
The connecting rod reference frame of each connecting rod of human skeleton model, is specifically included:
The connecting rod of the main joint containing two time joints for sub- joint, two axles of its reference frame respectively with two sons
Rotary shaft of the joint in initial joint angles is conllinear, and another rotary shaft is vertical with the plane that the two axles are formed, and is formed right
Hand rectangular coordinate system, in initial attitude, X-axis points to the front of human skeleton model, and Y-axis is pointing to human skeleton model just
Left, Z axis point to the surface of human skeleton model;
It is the connecting rod of the main joint containing one joint, a reference axis of its reference frame and sub- joint for sub- joint
Rotary shaft it is conllinear, a reference axis and the connecting rod bone collinear vector or vertical, another rotary shaft and the two axles are formed
Plane it is vertical, form right hand rectangular coordinate system, in initial attitude, X-axis points to the front of human skeleton model, and Y-axis refers to
To the front-left of human skeleton model, Z axis points to the surface of human skeleton model;
The origin of coordinates of link rod coordinate system is respectively positioned on connecting rod end point, i.e., at the sub- articulation center point of connecting rod.
8. a kind of anthropomorphic robot according to claim 1 imitates method for evaluating similarity, it is characterised in that:It is described by people
The connecting rod bone vector of body skeleton model is transformed into its female connecting rod reference frame, is specifically included:
Utilize transition matrix RmConnecting rod bone vector in depth camera coordinate system is transformed into the link rod coordinate system of its female connecting rod
In, transition matrix RmFor:
<mrow>
<msub>
<mi>R</mi>
<mi>m</mi>
</msub>
<mo>=</mo>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<mfrac>
<mrow>
<msub>
<mover>
<mi>x</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>&CenterDot;</mo>
<mover>
<mi>x</mi>
<mo>&RightArrow;</mo>
</mover>
</mrow>
<mrow>
<mo>|</mo>
<msub>
<mover>
<mi>x</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>|</mo>
<mo>|</mo>
<mover>
<mi>x</mi>
<mo>&RightArrow;</mo>
</mover>
<mo>|</mo>
</mrow>
</mfrac>
</mtd>
<mtd>
<mfrac>
<mrow>
<msub>
<mover>
<mi>x</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>&CenterDot;</mo>
<mover>
<mi>y</mi>
<mo>&RightArrow;</mo>
</mover>
</mrow>
<mrow>
<mo>|</mo>
<msub>
<mover>
<mi>x</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>|</mo>
<mo>|</mo>
<mover>
<mi>y</mi>
<mo>&RightArrow;</mo>
</mover>
<mo>|</mo>
</mrow>
</mfrac>
</mtd>
<mtd>
<mfrac>
<mrow>
<msub>
<mover>
<mi>x</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>&CenterDot;</mo>
<mover>
<mi>z</mi>
<mo>&RightArrow;</mo>
</mover>
</mrow>
<mrow>
<mo>|</mo>
<msub>
<mover>
<mi>x</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>|</mo>
<mo>|</mo>
<mover>
<mi>z</mi>
<mo>&RightArrow;</mo>
</mover>
<mo>|</mo>
</mrow>
</mfrac>
</mtd>
</mtr>
<mtr>
<mtd>
<mfrac>
<mrow>
<msub>
<mover>
<mi>y</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>&CenterDot;</mo>
<mover>
<mi>x</mi>
<mo>&RightArrow;</mo>
</mover>
</mrow>
<mrow>
<mo>|</mo>
<msub>
<mover>
<mi>y</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>|</mo>
<mo>|</mo>
<mover>
<mi>x</mi>
<mo>&RightArrow;</mo>
</mover>
<mo>|</mo>
</mrow>
</mfrac>
</mtd>
<mtd>
<mfrac>
<mrow>
<msub>
<mover>
<mi>y</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>&CenterDot;</mo>
<mover>
<mi>y</mi>
<mo>&RightArrow;</mo>
</mover>
</mrow>
<mrow>
<mo>|</mo>
<msub>
<mover>
<mi>y</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>|</mo>
<mo>|</mo>
<mover>
<mi>y</mi>
<mo>&RightArrow;</mo>
</mover>
<mo>|</mo>
</mrow>
</mfrac>
</mtd>
<mtd>
<mfrac>
<mrow>
<msub>
<mover>
<mi>y</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>&CenterDot;</mo>
<mover>
<mi>z</mi>
<mo>&RightArrow;</mo>
</mover>
</mrow>
<mrow>
<mo>|</mo>
<msub>
<mover>
<mi>y</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>|</mo>
<mo>|</mo>
<mover>
<mi>z</mi>
<mo>&RightArrow;</mo>
</mover>
<mo>|</mo>
</mrow>
</mfrac>
</mtd>
</mtr>
<mtr>
<mtd>
<mfrac>
<mrow>
<msub>
<mover>
<mi>z</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>&CenterDot;</mo>
<mover>
<mi>x</mi>
<mo>&RightArrow;</mo>
</mover>
</mrow>
<mrow>
<mo>|</mo>
<msub>
<mover>
<mi>z</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>|</mo>
<mo>|</mo>
<mover>
<mi>x</mi>
<mo>&RightArrow;</mo>
</mover>
<mo>|</mo>
</mrow>
</mfrac>
</mtd>
<mtd>
<mfrac>
<mrow>
<msub>
<mover>
<mi>z</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>&CenterDot;</mo>
<mover>
<mi>y</mi>
<mo>&RightArrow;</mo>
</mover>
</mrow>
<mrow>
<mo>|</mo>
<msub>
<mover>
<mi>z</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>|</mo>
<mo>|</mo>
<mover>
<mi>y</mi>
<mo>&RightArrow;</mo>
</mover>
<mo>|</mo>
</mrow>
</mfrac>
</mtd>
<mtd>
<mfrac>
<mrow>
<msub>
<mover>
<mi>z</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>&CenterDot;</mo>
<mover>
<mi>z</mi>
<mo>&RightArrow;</mo>
</mover>
</mrow>
<mrow>
<mo>|</mo>
<msub>
<mover>
<mi>z</mi>
<mo>&RightArrow;</mo>
</mover>
<mi>m</mi>
</msub>
<mo>|</mo>
<mo>|</mo>
<mover>
<mi>z</mi>
<mo>&RightArrow;</mo>
</mover>
<mo>|</mo>
</mrow>
</mfrac>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
WhereinIt is positive in the same direction with depth camera coordinate system X-axis, Y-axis, Z axis respectively in depth camera coordinate system
Vector,Respectively in depth camera coordinate system with the connecting rod referential coordinate of target connecting rod mother's connecting rod
The positive vector in the same direction of X-axis, Y-axis, Z axis;
Then for k-th of human skeleton model connecting rod for being used to calculate posture similarity, in the connecting rod referential of its female connecting rod
Connecting rod bone vectorFor:
<mrow>
<msup>
<mover>
<msub>
<mi>V</mi>
<mi>k</mi>
</msub>
<mo>&RightArrow;</mo>
</mover>
<mo>&prime;</mo>
</msup>
<mo>=</mo>
<msub>
<mi>R</mi>
<mrow>
<mi>m</mi>
<mi>k</mi>
</mrow>
</msub>
<mo>&times;</mo>
<mover>
<msub>
<mi>V</mi>
<mi>k</mi>
</msub>
<mo>&RightArrow;</mo>
</mover>
</mrow>
Wherein RmkFor corresponding transition matrix,For the connecting rod bone vector in depth camera coordinate system.
9. a kind of anthropomorphic robot according to claim 1 imitates method for evaluating similarity, it is characterised in that:The basis
Vectorial coordinate of each each connecting rod of joint angles calculating robot of robot in its female link rod coordinate system, is specifically included:
It is (0,0,1) to make initial attitude connecting rod vectorial coordinate of each connecting rod of robot in its female link rod coordinate systemT, it is female according to it
The anglec of rotation θ in each secondary joint in jointmki, utilize corresponding spin matrix R (θmki), obtain the connecting rod vectorial coordinate of current pose
<mrow>
<msup>
<mover>
<msub>
<mi>N</mi>
<mi>k</mi>
</msub>
<mo>&RightArrow;</mo>
</mover>
<mo>&prime;</mo>
</msup>
<mo>=</mo>
<munderover>
<mo>&Pi;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<mi>R</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>&theta;</mi>
<mrow>
<mi>m</mi>
<mi>k</mi>
<mi>i</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<mover>
<msub>
<mi>N</mi>
<mi>k</mi>
</msub>
<mo>&RightArrow;</mo>
</mover>
</mrow>
Wherein, n is the secondary joint number of target connecting rod female joint, and i numerical value is smaller to represent this joint in connecting rod-articulated chain
Closer to female connecting rod.
10. a kind of anthropomorphic robot according to claim 1 imitates method for evaluating similarity, it is characterised in that:Described
According to bone vectorial coordinate of the human skeleton model connecting rod in its female link rod coordinate system and robot links in its female connecting rod coordinate
Vectorial coordinate in system calculates the similarity of anthropomorphic robot and human body attitude, specifically includes:
According to the connecting rod vector after conversion coordinate, human skeleton model connecting rod bone vector sum corresponding machine people connecting rod vector is utilized
Angle cosine value calculate two connecting rod attitudes coefficient correlation, further calculate each group connecting rod coefficient correlation arithmetic average
Value, obtain the index of similarity of robot motion imitation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711202768.7A CN107818318B (en) | 2017-11-27 | 2017-11-27 | Humanoid robot simulation similarity evaluation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711202768.7A CN107818318B (en) | 2017-11-27 | 2017-11-27 | Humanoid robot simulation similarity evaluation method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107818318A true CN107818318A (en) | 2018-03-20 |
CN107818318B CN107818318B (en) | 2020-05-22 |
Family
ID=61610194
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711202768.7A Active CN107818318B (en) | 2017-11-27 | 2017-11-27 | Humanoid robot simulation similarity evaluation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107818318B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108762495A (en) * | 2018-05-18 | 2018-11-06 | 深圳大学 | The virtual reality driving method and virtual reality system captured based on arm action |
CN108961867A (en) * | 2018-08-06 | 2018-12-07 | 南京南奕亭文化传媒有限公司 | A kind of digital video interactive based on preschool education |
CN109064486A (en) * | 2018-06-20 | 2018-12-21 | 华南理工大学 | A kind of anthropomorphic robot and human body attitude method for evaluating similarity |
CN109432753A (en) * | 2018-09-26 | 2019-03-08 | Oppo广东移动通信有限公司 | Act antidote, device, storage medium and electronic equipment |
CN110135078A (en) * | 2019-05-17 | 2019-08-16 | 上海凌笛数码科技有限公司 | A kind of human parameters automatic generation method based on machine learning |
CN112634415A (en) * | 2020-12-11 | 2021-04-09 | 北方信息控制研究院集团有限公司 | Human body skeleton model-based real-time simulation method for human actions |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6980889B2 (en) * | 2003-10-08 | 2005-12-27 | Sony Corporation | Information processing apparatus and method, program storage medium, and program |
CN1830635A (en) * | 2005-03-10 | 2006-09-13 | 北京理工大学 | Human imitation robot action similarity evaluation based on human body motion track |
CN104816298A (en) * | 2015-05-12 | 2015-08-05 | 北京理工大学 | Similarity elevating method of humanoid robot tumble action |
CN106971050A (en) * | 2017-04-18 | 2017-07-21 | 华南理工大学 | A kind of Darwin joint of robot Mapping Resolution methods based on Kinect |
-
2017
- 2017-11-27 CN CN201711202768.7A patent/CN107818318B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6980889B2 (en) * | 2003-10-08 | 2005-12-27 | Sony Corporation | Information processing apparatus and method, program storage medium, and program |
CN1830635A (en) * | 2005-03-10 | 2006-09-13 | 北京理工大学 | Human imitation robot action similarity evaluation based on human body motion track |
CN104816298A (en) * | 2015-05-12 | 2015-08-05 | 北京理工大学 | Similarity elevating method of humanoid robot tumble action |
CN106971050A (en) * | 2017-04-18 | 2017-07-21 | 华南理工大学 | A kind of Darwin joint of robot Mapping Resolution methods based on Kinect |
Non-Patent Citations (2)
Title |
---|
HOSSEIN PAZHOUMAND-DAR,CHIOU-PENG LAM, MARTIN MASEK: ""Joint movement similarities for robust 3D action recognition using skeletal data"", 《J. VIS. COMMUN. IMAGE R. 30 (2015)》 * |
JIE LEI,MINGLISONG,ZE-NIANLI,CHUNCHEN: ""Whole-body humanoid robot imitation with pose similarity evaluation"", 《SIGNAL PROCESSING108(2015)》 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108762495A (en) * | 2018-05-18 | 2018-11-06 | 深圳大学 | The virtual reality driving method and virtual reality system captured based on arm action |
WO2019218457A1 (en) * | 2018-05-18 | 2019-11-21 | 深圳大学 | Virtual reality driving method based on arm motion capture, and virtual reality system |
CN109064486A (en) * | 2018-06-20 | 2018-12-21 | 华南理工大学 | A kind of anthropomorphic robot and human body attitude method for evaluating similarity |
CN108961867A (en) * | 2018-08-06 | 2018-12-07 | 南京南奕亭文化传媒有限公司 | A kind of digital video interactive based on preschool education |
CN109432753A (en) * | 2018-09-26 | 2019-03-08 | Oppo广东移动通信有限公司 | Act antidote, device, storage medium and electronic equipment |
CN110135078A (en) * | 2019-05-17 | 2019-08-16 | 上海凌笛数码科技有限公司 | A kind of human parameters automatic generation method based on machine learning |
CN112634415A (en) * | 2020-12-11 | 2021-04-09 | 北方信息控制研究院集团有限公司 | Human body skeleton model-based real-time simulation method for human actions |
CN112634415B (en) * | 2020-12-11 | 2023-11-10 | 北方信息控制研究院集团有限公司 | Human body skeleton model-based personnel action real-time simulation method |
Also Published As
Publication number | Publication date |
---|---|
CN107818318B (en) | 2020-05-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107818318A (en) | A kind of anthropomorphic robot imitates method for evaluating similarity | |
CN106078752B (en) | A kind of anthropomorphic robot human body behavior imitation method based on Kinect | |
CN104598867B (en) | A kind of human action automatic evaluation method and dancing points-scoring system | |
CN107953331A (en) | A kind of human body attitude mapping method applied to anthropomorphic robot action imitation | |
CN109064486A (en) | A kind of anthropomorphic robot and human body attitude method for evaluating similarity | |
CN112069933A (en) | Skeletal muscle stress estimation method based on posture recognition and human body biomechanics | |
CN107598897A (en) | A kind of method of humanoid robot gait's planning based on human body teaching | |
Ayusawa et al. | Real-time implementation of physically consistent identification of human body segments | |
CN107330967A (en) | Knight's athletic posture based on inertia sensing technology is caught and three-dimensional reconstruction system | |
CN109544666A (en) | A kind of full automatic model deformation transmission method and system | |
CN203673431U (en) | Motion trail virtual device | |
CN104678766B (en) | A kind of optimal batting acquiring method of configuration of apery mechanical arm flight spheroid operation | |
CN106607910A (en) | Robot real-time simulation method | |
CN108098780A (en) | A kind of new robot apery kinematic system | |
CN103824326A (en) | Dynamic human body three-dimensional modeling method | |
CN107616898A (en) | The wearable healing robot of upper limbs and Rehabilitation Assessment method based on everyday actions | |
CN108447077A (en) | A kind of horsemanship jockey posture information acquisition analysis system | |
CN113158459A (en) | Human body posture estimation method based on visual and inertial information fusion | |
CN102156994B (en) | Joint positioning method for single-view unmarked human motion tracking | |
CN103729879A (en) | Virtual hand stable grabbing method based on force sense calculation | |
CN117697717A (en) | Exoskeleton physical man-machine two-way interaction simulation system | |
Chakravarthi et al. | Real-time human motion tracking and reconstruction using IMU sensors | |
Kondo et al. | Real-time posture imitation of biped humanoid robot based on particle filter with simple joint control for standing stabilization | |
Nicolas et al. | From bone to plausible bipedal locomotion using inverse kinematics | |
Nagarsheth et al. | Modeling and dynamics of human arm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |