SlideShare une entreprise Scribd logo
1  sur  51
생체 이해에
기반한 로봇
김용재 Ph.D.
IRIM Lab. Koreatech
고성능 로봇에게
인간의 유연함과
안정성 부여하기
PROFESSIONAL EXPERINECE
2014.3 - Present Assistant Professor, KoreaTech
2012.3 - 2014.3 Research Staff Member, Intelligent Robot Group, SAIT
2011. 3 - 2012.2 Principal Engineer and Visiting Engineer of MIT
Manufacturing Tech. Center, Samsung Electronics
2003. 3 - 2011.2 Senior Engineer, Robot Technology Team
Manufacturing Tech. Center, Samsung Electronics
Education :
~2003, Ph.D in KAIST
~1998, M.S in KAIST
~1996, B.S in KAIST
Major:
Electrical Engineering
and Computer Science
Yong Jae Kim, Ph.D.
RESEARCH INTERESTS
- Mechanism design and Control of
Wearable Robots, Surgical Robots, Humanoid & Interactive Mobile Robots
- Design and Analysis of Flexible Robots and Backdrivable Mechanisms
- Compliance Control for Backdrivable Mechanism & Redundant manipulators
- Idea Generation of Novel Robot System (22 US patents, TRIZ expert Level II)
※ Ref: ABB Yumi
Personal
Robot
Public Robot
“iMaro-PSR”
Humanoid
“RoboRay”
Single Port
Surgical Robot
Samsung
Cleaning robot
2004 2005 2006 20102007 2009 2012 2013
Samsung Robot
Commercialization TF
Cobot
“Rayarm”
Walking Assist
Robot
SAMSUNG ELECTRONICS (2003-2013)
Flexible Human
Assist Robot (Confidential Project)
(2014)
Compliant Gripper (미래부)
•2016~
Compression
support frame
Kevlar
Belt
Ni-Ti Cable
Non-stretchable
Fabric
Conduit
Power Assist Glove
•2016~2004 2005 2006 2015 2016 2017
High-Payload
Gravity Compensation
Manipulator
(Confidential Project)
Torque sensing
Manipulator (Confidential Project)
•2015~2004 2005 2006 20152014 2016
DARPA Robotics
Challenge (산자부)
(2014-15)
LIMS: Highly Interactive
Manipulator (2014 ~) Anthropomorphic
dual arm robot for
High speed Interaction
(LAVER LABS)
•2009
•2012 •2013
•2010
2012 2014 2015 2016 20172010
Interactive Robotics &
Innovative Mechanism LabIRIM
•2014
KOREATECH(2014-2017)
TABLE OF CONTENTS
1_ 인간의 몸과 로봇 메커니즘 “산업용 로봇은 왜 박수를 칠 수 없을까?”
2_ 생체 이해에 기반한 로봇 핸드 “유연성과 정밀성을 모두 갖춘 로봇이 가능할까?”
2.1_ RoboRay Hand: 고자유도 휴머노이드 핸드
2.2_ TATH Glove : 파지력 증강 소프트 글러브
2.3_ Soft Parallel Gripper: 정밀 유연 그리퍼
3_ 생체 이해에 기반한 로봇 팔 “하이 파이브가 가능한 로봇의 조건은?”
3.1_중력보상 로봇, 가변강성 로봇
3.2_ LIMS2–AMBIDEX: 경량 고강성 로봇 팔
4_ 인간과 협력하는 로봇
1.인간의 몸과 로봇 메커니즘
“산업용 로봇은 왜 박수를 칠 수 없을까?”
Robot Manipulators vs. Human Arms
Weight :
Joint speed :
TCP speed :
Payload :
Rotational inertia :
Stiffness :
Repeatability :
Control frequency :
23.9 kg
90 ~180 deg/sec
Approx. 1.5m/sec
7 kg
over 5kgm2
10,000~40,000 Nm/rad
0.1 mm
1kHz ( 3~4kHz, joint control)
Weight :
Joint speed :
TCP speed :
Payload :
Rotational inertia :
Stiffness :
Repeatability :
Control frequency :
3.9 kg
400 ~ 1200 deg/sec
over 15m/sec
-
0.49kgm2
350 Nm/rad (elbow)
?
10Hz (40Hz, uncond. reflex)
Video Ref: ABB Robotics - Fanta Can Challenge- Level II - Superior Motion Control
https://www.youtube.com/watch?v=SOESSCXGhFo
Video Ref: Last 5 Minutes of Super Bowl 51 https://www.youtube.com/watch?v=iN8Rfwr6L2U
Pull Up & Chin Up Progression Guide incl. 10+ Exercises https://m.youtube.com/watch?v=CdtrfXK7bcg
Men's Archery Individual Gold Medal Match | Rio 2016 Replay https://www.youtube.com/watch?v=rzj4FFi7wt8
Robot Grippers vs. Human Hands
Weight :
Degrees of Freedom :
Joint speed :
Fingertip force :
Grasp force :
No. of Muscles :
1.5 kg
23 DOF
Over 500 deg/sec
43 N (index, straight pose)
370 N
over 30 muscles
Video Ref.: Grasping different objects quickly https://www.youtube.com/watch?v=ZNjbyCWt93k
Flexible Robot Gripper: 2-finger Adaptive Electric Robot Gripper by Robotiq https://www.youtube.com/watch?v=nkGuI4uiSLM
Soft Robotics Inc. - Adaptive Grasping of Common E-Commerce Items https://www.youtube.com/watch?v=vRQZ5bqwDbg
Video Ref.: Barbell Deadlift Bent Row Complex https://www.youtube.com/watch?v=PUNxkzCjWNk
World's most complicated watch https://www.youtube.com/watch?v=I1L15xehfEA
Slow Motion Rubik's Cube Solve https://www.youtube.com/watch?v=gEPgcWx1JMs
Contact materials?
Sensors?
Control Algorithm?
…
Robot Grippers vs. Human Hands
Actuation
- More than 30 muscles
- Highly redundant!
- Tendon driven
Actuation
- 1~ 16 motors
- Difficult to
increase DOF.
- Gear or wire
driven
인체는 구동기의
증가에 대한 부담
이 없다!!
Lubrication
- Lub. by Synovial fluid
- Friction coeff.: 0.003~0.01
Lubrication
- Bearing Friction coeff.:
approx. 0.026
- Teflon : 0.04
베어링 대비
2.6~8.6 분의 1
저마찰 윤활유가
평생 공급된다!!
Contact materials?
Sensors?
Control Algorithm?
…
생체 모양만 모사하는 것은 무의미하다! 어느 level에서 모사해야 할까?
Joints
- Rolling & sliding
- Aligned by ligaments
Joints
- Revolute joints
- Bearing for low frictions
- Vulnerable to Impact
Maintenance
Free!!
평생 장력이 유지
되는 인대구조.
인간의 능력을 모사하는 로봇들
Precise,
High Payload
High Speed
Soft/flexible,
Interactive,
Compliant
유연한 로봇은 정밀한 로봇이 될 수 없는가? 인간은 어디에 위치하는가?
Industrial Robot
+ force sensors
+ contact detection
+ …
Soft Robot
+ Increase payload
+ Position sensors
+ …
2. 생체 이해에 기반한 로봇 핸드
“유연성과 정밀성을 모두 갖춘 로봇이 가능할까?”
C6M
Shadow Robotics
3-Finger Gripping Hand
Schunk
5-Finger Hand
Schunk
i-Limb Ultra
Touch Bionics
Robonaut hand
NASA
Barrett hand
Barrett Corp.
Human Hand
27 Bones, 23 DOF, Over 30 muscles
Industrial
Prosthetic
Research & General purpose
C6M
DLR
RoboRay hand
Samsung
Humanoid
The goal : Precise Manipulation + Compliant Power Grasping
인간과 유사한 크기/모양/무게, 마찰 최소화 및 역구동성 최대화
RoboRay Hand ( Samsung Project, published in ICRA2014 )
RoboRay Hand : Highly backdrivable Life-Size Robotic Hand
– 1.59kg, 12 DOF for fingers and 2 DOF for the wrist
– Fingertip force 15N / Contact force detection 0.196N
– Tendon-driven mechanism actuated by ball screws and BLDC motors
Weight 1.59kg (including wrist and forearm)
Dimensions
Hand 160 x 80 x 45 mm
Forearm 186 x 76 x 83 mm
DOF
Fingers 12 DOF / 5 Fingers
Wrist 2 DOF
Payload
Peak Fingertip
Force
15N (Stretched Pose)
Speed/
Reduction Ratio
MPR joint 800deg/sec, 47:1
MPP joint 700deg/sec, 57:1
PIP joint 450deg/sec, 82:1
Sensing
Minimum Force Detection
with compensation : 0.196N
w.o. compensation : 0.735N
Actuation
Tendon driven, actuated by
ball screws and BLDC Motors
Electronics
DC 12V, DSP TMS320F2812
EtherCAT Communication
2.1 RoboRay Hand- Specification ( Samsung Project, published in ICRA2014 )
RoboRay Hand
- Samsung Project
- Published in
ICRA2014
“RoboRay Hand : A Highly
Backdrivable Robotic Hand
with Sensorless Contact Force
Measurements”
Yong-Jae Kim, et al, ICRA2014
Orangutan Baboon Human RoboRay
40 57-58 60 App. 100
Opposability among primates
Metacarpus
of the thumb
1. Biomimetic Design based on functionality of human hands
2.1 Mechanical Design - Configuration
High Opposability
– Human thumb is highly opposable
due to mobile metacarpus
Finger Design Considering Opposability
2. Actuator Configuration Based on the Grasping Functions
Precise Manipulation
Grasp motor
fixed
Pose motor
in motion
- Fixing the grasp motor and actuating
pose motor enable precise adjustment.
Compliant Grasping
Pose
motor
off
Grasp motor
In motion
- Turning the pose motor off and actuating
the grasp motor make compliant grasping
High force
grasp motor
in the forearm
Small
Pose Motor
in the Palm
Proposed
Configuration
MP
joint
PIP
joint
DIP
joint
Underactuated Hands
- Inherent mechanical compliance
- Difficult to manipulate object precisely
Dexterous Hands
- Fully controllable, Precise manipulation
- Subject to be bulky, or slow and weak
2.1 Mechanical Design – Finger Actuation Concept
3. Tension Decoupling Wrist Mechanism using Rolling Joint
Transmission using Conduits
- Saving space, High force
- Frictional, Limited Range of Motion
Built-in Actuators in the Palm
- Relatively Simple
- Subject to be bulky, weak or slow
Proposed
Wrist Mechanism
Wrist Pitch
Motion
Wrist Roll
Motion
Pitch
Direction
Roll
Direction
2p
1p
r
Rolling Joint for Wrist Pitch
Offset Pivot for Wrist Roll
Composed 2-DOF Joint
Rolling
Joint
2.1 Mechanical Design - Wrist Decoupling Concept
Stand-alone Performance Test
Gestrure and Grasping Test
(Prototype of RoboRay hand)
Gesture and Manipulation Test
(RoboRay hand without Covers)
Performance back-drivability
- backdrivable force : 0.735N
- Mechanical Efficiency : App. 89.5%
- Stiffness : App. 1.7N/mm
Bearing Efficiency 97.0%
No. Radius (mm) Angle (deg) Efficiency (%)
1 3.2 90 99.4
2 4.0 180 99.4
3 10.0 180 99.7
4 10.0 180 99.7
5 4.0 90 99.5
6 4.0 180 99.4
7 10.0 90 99.8
Ball Screw Efficiency 95%
Actuator Bushing Efficiency 95%
Total Efficiency 89.4%
2.1 Mechanical Design - Wrist Decoupling Concept
Flexible/Soft
and Safe
Rigid, Precise
and Strong
Human
hand
- 유연성 + 정밀성
- 기능 수준의 생체 모사
- 기존의 기계적인 요소품 사용
2.1 RoboRay Hand
2.1 Flexible/Soft but Strong?
Flexible/Soft
and Safe
Rigid, Precise
and Strong
Human
hand
Flexible/Soft + High payload
?
- 유연한 구조인 동시에
강한 힘을 전달할 수 있다면...
FDC Structure (Force Distributing Compliant Structure)
- 말단부의 강성을 높이거나 조절 가능
- 구조에 따라 규제되는 자유도를 선택가능
Vertical & Lateral
Motion Restricted
torsional motion
permitted
Fin-Gripper
(FESTO. Co)
Vertical Motion
Restricted
Lateral & torsional
motion permitted
Compliant
to side force
Vertical & torsional
Motion Restricted
Lateral motion
permitted
It can withstand
Vertical & torsional Loads!
기존 구조물 Cantilever Beam
- Stiff proximal part
- Soft distal part
High
Stiffness
Low
Stiffness
2.2 New Opportunities by FDC structure
It can withstand Vertical,
lateral & torsional Loads!
All 3 motions
restricted
단순 Cantilever Beam
Flexible
beam
Flexible
beam
Dovetails
for
prismatic
motion
Walking Assist Robot, GEMS
Younbaek lee, IROS2017
SAIT, Samsung electronics
GEMS (SAIT, Samsung)
Small
deformationF
Large
deflection
F
Fin-gripper,
Festo.
Fin-gripper (FESTO Co.)
2.2 New Opportunities by FDC structure
FDC Structure
Fin-Gripper
(FESTO. Co)
tension
Except finger bones, all other parts
receive tensile force!
FDC Structure
in Human Fingers
The FDC structure for one-side force direction and curvature can be composed only using soft
fabric and belt except one flexible frame.
The proposed structure facilitates transfer of tactile sensation because there are soft fabric and belt
between the object and human hand.
Tension
Compre
-ssion
Tension
Tension
- Only the right flexible frame receives compressive force.
- The others can be replaced by soft materials!
Human finger case
Tension
Moment Assist
Flexible frame
(ABS+NiTi wire)
Belt
(Kevlar)
Non-
stretchable
Fabric
(Polyester)
Only soft materials
between the object
and the finger!!
Wearable FDC Structure
for the Moment Assisting
Glove
2.2 FDC Structures for Moment Assisting Devices
TActily Transparent High-Force Assist Glove (TATH Glove) : 촉감 전달이 가능한 파지력 증강 글러브
2-Finger Test Mechanism
Palm side
of finger
Compression
support frame
Conduit
Kevlar
Belt
Nitinol Wire
Low-stretchable
Fabric
Conceptual Design of
5-Finger TATH Glove Concept Proof Mechanism Loading Test (7kg)
2.2 TATH Glove using FDC Structure
Customized
FSR Sensor
2.2 TATH Glove using FDC Structure
전기전자통신공학부
Interactive Robotics &
Innovative Mechanism Lab
2.1 Flexible/Soft but Strong?
Flexible/Soft
and Safe
Rigid, Precise
and Strong
Human
hand
Flexible/Soft + High payload
?
Flexible/Soft
and Safe
Rigid, Precise
and Strong
Human
hand
Precise parallel
pinching
Parallel pinching
Compliant Grasping
손가락은 부드럽게 감아 쥐지만
손끝은 평행한 정밀파지가 가능?
Compliant
grasping
2.2 Precise parallel pinching + compliant grasping?
Parallel pinching
Motion
Compliant Grasping
Motion
Fingertip은 수평을 유지하는 동시에, finger body는
flexible한 구조로 물체를 감싸는 동작이 가능함Finger body
유연 파지
Fingertip
평행 핀칭
Triangular
structure
Parallel
structure
2.2 Gripper with Precise Parallel Pinching & Compliant Grasping
Soft Parallel Gripper : 정밀 평행 파지가 가능한 유연 그리퍼
생체 이해에 기반한 로봇 핸드 -로봇이 유연성과 정밀성을 모두 가지려면
• 생체 이해에 기반한 로봇 핸드
– RoboRay Hand : 정밀조작과 유연 고하중 파지가 가능한 로봇 핸드
– THAT Glove : 촉감전달이 가능한 파지력 증강 글러브
– Soft Parallel Gripper : 정밀 평행 파지가 가능한 유연 그리퍼
• 생체모사는 형상모사가 아니다. 원리와 기능 레벨에서의 생체 모사가 필요하다
• 유연성과 고하중/정밀성은 trade-off 가 아니다. 양쪽을 다 만족하는 구조 도출
Organic
mechanism
Fundamental
principle
Robot Application Organic
mechanism
Fundamental
principle Robot Applications
전기전자통신공학부
Interactive Robotics &
Innovative Mechanism Lab
3. 생체 이해에 기반한 로봇 팔
“하이 파이브가 가능한 로봇의 조건”
Ave. Inertia : 0.49kgm2
Ave. Mass : 3.9kg
Estimated Inertia : over 5kgm2
Mass : 23.9kg
Kinetic and potential energy : make a big difference to the safety !!
Stiffness/strength : critical to control performance !!
인간의 팔과 로봇 팔의 안전성과 성능 차이
High-five motion
7.13 m/sec
Throwing Motion
15.61 m/secPrecise, Strong, but dangerous Inherently safe and still strong, precise and fast
VS.
Ev = mv2
1
2
Ep = mgh
운동에너지
위치에너지
높은 강도/강성/정밀도  높은 제어 성능 낮은 질량, 높은 역구동성 근본적인 안정성
Design Philosophy and Basic Idea
Motors with
gears
in base frame
Tendons for
joint actuation
Reduction
gears
Motors
Motors
in base frame
Reduction
gears
Tendons for
joint actuation
Conventional Mechanism : Stiff but Heavy
Tendon Driven Mechanism : light weight but low stiffness
Hybrid mechanism : light and Stiff,
light-weight gear mechanism is required!
1) Extremely low inertia to minimize stored kinetic energy.
2) Extremely low mass. Conventional industrial robots consume
most of the motor torque for supporting their own weight.
3) High stiffness comparable to industrial robots.
4) High strength comparable to industrial robots.
5) Efficiency and backdrivability.
In order to apply right amount of energy to the robot, efficient
mechanism with minimal friction is required. It enables sensing of
external force without expensive force or torque sensors.
Stiffness
350Nm/rad
Ave. Inertia : 0.49kgm2
Ave. Mass : 3.9kg
근본적인 안전성과 높은 제어성능을 가지기 위한 조건
170 mm 340 mm370 mm
3D-Printed
ABS frames
Aluminum alloy,
machining parts
인간과 동일한 7 자유도 팔
유연파지가 가능한 핸드
3-DOF
Wrist
1-DOF
Elbow
3-DOF
Shoulder
4-DOF
Hand
3-DOF Neck
Pan, Tilt and Translation
4 Actuators
for Wrist and
Elbow
3 Actuators with
Capstan Reduction
for shoulder
무거운 구동모터들
Body 근처에 배치
Joints with
Tension Amplification
Mechanisms
강성을 증폭하는
경량 감속 메커니즘
LIMS2-AMBIDEX
( Low Inertia Manipulator with High Stiffness and Strength )
OVERALL MOTION
-Arm span 2.1m
-7 DOF per arm
-3 DOF per head
-Wrist Yaw ROM:
670 deg
35
ARM WEIGHT
Total : 2,628 g
Wrist : 446 g
Forearm : 171 g
Elbow : 533 g
Upper arm:1,478 g
HIGH SPEED
MOTION TEST
Strength and Stiffness Analysis (LIMS1 Results)
z
Stiffness:
Stiffness/Inertia:
>

10,000 Nm/rad
8,333 m/rad
1,410 Nm/rad
7,233 m/rad
350 Nm/rad
2,059 Nm/rad
>
>
• 산업용 로봇과 유사한 동적 제어 성능 – 강성/질량
% Anthropomorphic Low-Inertia High-
Stiffness Manipulator for High-Speed Safe
Interaction, IEEE Trans. on Robotics, 2017
• 반복 정밀도 0.426mm (산업용 로봇 0.1mm)
P2P3
P4 P5
P1
500×500×500mm
(a) (b)
P2
P3
P4
P5
P1
TABLE V
POSITIONING REPEATABILITY TEST RESULTS
Pose
Mean
(mm)
Std. Dev.
(mm)
3-Sigma
(mm)
P1 0.138 0.066 0.335
P2 0.148 0.074 0.369
P3 0.103 0.048 0.247
P4 0.187 0.113 0.528
P5 0.189 0.116 0.536
Total 0.153 0.091 0.426
TABLE V
POSITIONING REPEATABILITY TEST RESULTS
Pose
Mean
(mm)
Std. Dev.
(mm)
3-Sigma
(mm)
P1 0.138 0.066 0.335
P2 0.148 0.074 0.369
P3 0.103 0.048 0.247
P4 0.187 0.113 0.528
P5 0.189 0.116 0.536
Total 0.153 0.091 0.426
• Payload 3kg
(a) (b)
0 10 20 30 40 50
0
100
200
300
400
500
600
700
800
900
1000
HIC
36
(m
robot
)
Effective mass of robots (kg)
HIC36
1m/sec
2m/sec
3m/sec
4m/sec
5m/sec
6m/sec
7m/sec
8m/sec
9m/sec
10m/sec
0 2 4 6 8 10
2
3
4
5
6
7
8
9
10
Effective mass (kg)
Speed(m/sec)
HIC36= 100
HIC36= 200
HIC36= 300
HIC36= 500
HIC36= 1000
1.47kg
• HIC (Head Injury Criteria) :
a safety criteria derived from
the average acceleration of a
human head and the
application time. mmotor n : 1 m1ink
k khead
mhead
Motor side mass Reducer Link side mass
Under 2m/sec, mass is not a dominant factor of safety ( Any mass satisfy HIC100).
At 6m/sec, effective mass 10kg can cause almost HIC1000 !
To be safe (HIC100) up to 5m/sec, the effective mass must be under 1.47kg !















 
5.2
12
12
,
2
121
1
)(max dtx
tt
ttHIC
t
t
H
tt

HIC 100 : non-life-threatening to the brain,
HIC 1000 : life-threatening injuries
Influence of Mass and Speed to the Safety
10kg, 6m/sec
HIC 958 !!
10kg, 2m/sec
HIC 61
Ev = mv2
1
2
0 10
0
100
200
300
400
H
100
200
105
(a)
0 10 20 30 40 50
0
100
200
300
400
500
600
700
800
900
1000
HIC
36
(m
robot
)
Effective mass of robots (kg)
HIC36
1m/sec
2m/sec
3m/sec
4m/sec
5m/sec
6m/sec
7m/sec
8m/sec
9m/sec
10m/sec
0
2
3
4
5
6
7
8
9
10
Speed(m/sec)
1.1kg, 6m/sec
HIC 100
INTERACTION
TEST
: high-five
motion
HIGH SPEED
IMPACT TEST
: Clapping
• Inertia & Stiffness
– Shoulder to Hand Inertia : 0.502 kgm2
– Elbow Stiffness : 1,410 Nm/rad
• Speed & peak torque
– Shoulder roll pitch: 499deg/s, 42.5~82Nm
– Shoulder Yaw: 749deg/s, 28.4~54.6Nm
– Elbow : 590deg/s, 69.4Nm
– Wrist roll pitch: 1179deg/s, 34.7Nm
– Wrist yaw: 1634deg/s, 25Nm
Specifications
Stiffness
350 Nm/rad
Inertia : 0.49 kgm2
Mass : 3.9 kg
Stiffness
10,000 Nm/rad
Estimated Inertia : 5 kgm2
Mass : 23.9 kg
Inertia : 0.50 kgm2
Mass : 2.63 kg
Stiffness
1,410 Nm/rad
AMBIDEX - LIMS2
Human Arm
Industrial Robot (iiwa)
• Degrees of Freedom
– 7 DOF / Arm
Shoulder 3, elbow 1, wrist 3
– 3 DOF for Neck
Pan, tilt and translation
• Weight
– Arm moving part : 2.63 kg
– Shoulder : 4.17 kg
– Hand : 0.633 kg
LIMS1
LIMS2-AMBIDEX
• Block and Tackle - light weight reduction mechanism
1-DOF Tension Amplification Joint
Motor for
Wire Motion
Rolling Joint
Spring Coefficient
of wire K
Wire Tension
Tin
Output Tension
Tout
Actuator Block and Tackle
outxinx
,inout nTT  inout x
n
x 
1
in
in
in
out
out
out Kn
x
T
n
x
T
K 22





Amplified
torque
Reduced
motion
Motor for
Wire Motion
Motor for
Wire Motion
Motor for
Wire Motion
Pulley
– Symmetric wire motion
– High Friction
– Limited Range of motion
• Tension amplification mechanism for revolute joints
Strength : amplified by n times!
Stiffness : amplified by n2 times!
– Symmetric wire motion
– Low Friction
– Wide Range of motion
※“An Innovative Trans-umbilical
Single-Port Surgical Robot” ICRA2014
z 2


w
w
leftl
rightl
d
• Joint Design
Elbow Joint using the 1-DOF Tension Amplification Joint
Relationship between
wire motion & joint angle
Developed Elbow Joint
n=6  Stiffness is 36 times higher!
Wire pair for
wrist actuation
Wrist wire length is decoupled
with the wrist motion!!
2
sin

nwll rightleft 
Symmetric
wire motion!
2

n
lleft
baseleftl _
baseleftl _
2
leftl
2
leftl
al
bl
cl
dl
prpr
Simplified and actual wire path
)2( dactualsimple lrll  
Constant
The same motion!
for even number of winding
2-DOF Tension Amplification Joint
Two 1-DOF
Joints
– Suitable for miniaturization
– Frictional wire transmission
– Limited workspace due to
extended wrist
※“An Innovative Trans-umbilical Single-Port
Surgical Robot” ICRA2014
Serial Connection of the 1-DOF tension amplification joint
– How to realize spherical
rolling contact without
slip
– Symmetricity of 2 wire
pairs should be verified
Extension to 2-DOF Mechanism
1-DOF Joint 2-DOF Concept
Challenging points
z
1-DOF Concept
2-DOF Tension Amplification Joint
z
Link 1
Link 2
d
leftPl _
rightTl _
rightPl _
leftTl _
w
a

sinw
cosw

sinw
cosw
b
a b
a b
Bending
Plane
Symmetric
relationship
between wires
2
sincos
2
sinsin
__
__




nwll
nwll
rightTleftT
rightPleftP


Motion of each wire of the wire pairs
is symmetric!
Link 1
2p
1p
2t
1t
distp
Link 2
proxp
Ideal Model
0
Link 1
Link 2
1p
2p
2t
1t
distp
proxp
LIMS1 : 구면 rolling 모션 모사 LIMS2 : 3개 link로 단순화, 내구성 향상
Link Type Rolling Joint를 이용한 wrist 개선
Wire for wrist actuation :
n=4 Stiffness is 16 times higher!
Cross roller
Planetary gear
Universal Joint
Wrist 말단부 관절 동력전달 및 감속 구조
• 3-finger 4-DOF 핸드
– 파지 물체에 따라 적절히 변형하는 underactuation 구조
– Thumb 자세를 변경하는 추가자유도로 grasping  pinching 변환 가능
LIMS Hand 개발
안전한 로봇 메커니즘 + 
괜찮아
들어와..
정말로
안 아파… 응?
어쩐지 어색해.
Summary and Future Work
인간과 협력하는 로봇
Summary and Future Work
• 생체 이해에 기반한 로봇 팔 “하이 파이브가 가능한 로봇의 조건은?”
– LIMS2–AMBIDEX: 안전성과 높은 제어 성능을 가진 로봇 팔
• 생활 속에 로봇이 들어오려면 …
Video Ref.: How Smartphones Are Assembled & Manufactured In China. https://www.youtube.com/watch?v=gBL-u53sy_o&t=1s
2013 Artistic Gymnastics World Championships https://www.youtube.com/watch?v=O4pSYZNFej0
The Best Female Rock Climber In the World is 14 Years Old https://m.youtube.com/watch?v=D4zBVD0sL7Y
전기전자통신공학부
Interactive Robotics &
Innovative Mechanism Lab
Thank you for your attention!
김종인
전형석
정용준
이덕원
장우석
Interactive Robotics &
Innovative Mechanism Lab

Contenu connexe

Tendances

THiRA-UTECH x THiRAbot
THiRA-UTECH x THiRAbotTHiRA-UTECH x THiRAbot
THiRA-UTECH x THiRAbotTHiRA-UTECH
 
Seminar report on captcha
Seminar report on captchaSeminar report on captcha
Seminar report on captchakunalkiit
 
NDC15 - 사례로 살펴보는 MSVC 빌드 최적화 팁
NDC15 - 사례로 살펴보는 MSVC 빌드 최적화 팁NDC15 - 사례로 살펴보는 MSVC 빌드 최적화 팁
NDC15 - 사례로 살펴보는 MSVC 빌드 최적화 팁Yi-kwon Hwang
 
금융 데이터 이해와 분석 PyCon 2014
금융 데이터 이해와 분석 PyCon 2014금융 데이터 이해와 분석 PyCon 2014
금융 데이터 이해와 분석 PyCon 2014Seung-June Lee
 
NDC2012_마비노기 영웅전 카이 포스트모템_시선을 사로잡는 캐릭터 카이 그 시도와 성공의 구현 일지
NDC2012_마비노기 영웅전 카이 포스트모템_시선을 사로잡는 캐릭터 카이 그 시도와 성공의 구현 일지NDC2012_마비노기 영웅전 카이 포스트모템_시선을 사로잡는 캐릭터 카이 그 시도와 성공의 구현 일지
NDC2012_마비노기 영웅전 카이 포스트모템_시선을 사로잡는 캐릭터 카이 그 시도와 성공의 구현 일지영준 박
 
Introduction to robotics
Introduction to roboticsIntroduction to robotics
Introduction to roboticsBikash Shrestha
 
Humanoid robot
Humanoid robotHumanoid robot
Humanoid robotSom Mishra
 
Profiling - 실시간 대화식 프로파일러
Profiling - 실시간 대화식 프로파일러Profiling - 실시간 대화식 프로파일러
Profiling - 실시간 대화식 프로파일러Heungsub Lee
 
Humanoid robotics
Humanoid roboticsHumanoid robotics
Humanoid roboticsGopal Verma
 
[26]자동화, 계륵에 살 붙이기 : Evolution of Android Automation Test
[26]자동화, 계륵에 살 붙이기 : Evolution of Android Automation Test[26]자동화, 계륵에 살 붙이기 : Evolution of Android Automation Test
[26]자동화, 계륵에 살 붙이기 : Evolution of Android Automation TestNAVER Engineering
 
Humanoid Service Robots: Challenges and Opportunities in Healthcare
Humanoid Service Robots: Challenges and Opportunities in HealthcareHumanoid Service Robots: Challenges and Opportunities in Healthcare
Humanoid Service Robots: Challenges and Opportunities in HealthcareSelcen Ozturkcan
 
2015 py con word2vec이 추천시스템을 만났을때
2015 py con word2vec이 추천시스템을 만났을때 2015 py con word2vec이 추천시스템을 만났을때
2015 py con word2vec이 추천시스템을 만났을때 choi kyumin
 
Intelligent mobile Robotics & Perception SystemsIntelligent mobile Robotics ...
Intelligent mobile Robotics  & Perception SystemsIntelligent mobile Robotics ...Intelligent mobile Robotics  & Perception SystemsIntelligent mobile Robotics ...
Intelligent mobile Robotics & Perception SystemsIntelligent mobile Robotics ...Gouasmia Zakaria
 
Lecture 1 - Introduction to Robotics.pptx
Lecture 1 - Introduction to Robotics.pptxLecture 1 - Introduction to Robotics.pptx
Lecture 1 - Introduction to Robotics.pptxSelvamM36
 

Tendances (20)

THiRA-UTECH x THiRAbot
THiRA-UTECH x THiRAbotTHiRA-UTECH x THiRAbot
THiRA-UTECH x THiRAbot
 
Artificial intelligence
Artificial intelligenceArtificial intelligence
Artificial intelligence
 
Seminar report on captcha
Seminar report on captchaSeminar report on captcha
Seminar report on captcha
 
NDC15 - 사례로 살펴보는 MSVC 빌드 최적화 팁
NDC15 - 사례로 살펴보는 MSVC 빌드 최적화 팁NDC15 - 사례로 살펴보는 MSVC 빌드 최적화 팁
NDC15 - 사례로 살펴보는 MSVC 빌드 최적화 팁
 
금융 데이터 이해와 분석 PyCon 2014
금융 데이터 이해와 분석 PyCon 2014금융 데이터 이해와 분석 PyCon 2014
금융 데이터 이해와 분석 PyCon 2014
 
NDC2012_마비노기 영웅전 카이 포스트모템_시선을 사로잡는 캐릭터 카이 그 시도와 성공의 구현 일지
NDC2012_마비노기 영웅전 카이 포스트모템_시선을 사로잡는 캐릭터 카이 그 시도와 성공의 구현 일지NDC2012_마비노기 영웅전 카이 포스트모템_시선을 사로잡는 캐릭터 카이 그 시도와 성공의 구현 일지
NDC2012_마비노기 영웅전 카이 포스트모템_시선을 사로잡는 캐릭터 카이 그 시도와 성공의 구현 일지
 
Humanoid robot
Humanoid robotHumanoid robot
Humanoid robot
 
human robot interaction
human robot interactionhuman robot interaction
human robot interaction
 
Introduction to robotics
Introduction to roboticsIntroduction to robotics
Introduction to robotics
 
AI
AIAI
AI
 
Humanoid robots
Humanoid robotsHumanoid robots
Humanoid robots
 
Humanoid robot
Humanoid robotHumanoid robot
Humanoid robot
 
Profiling - 실시간 대화식 프로파일러
Profiling - 실시간 대화식 프로파일러Profiling - 실시간 대화식 프로파일러
Profiling - 실시간 대화식 프로파일러
 
Humanoid robotics
Humanoid roboticsHumanoid robotics
Humanoid robotics
 
Humanoid robotics
Humanoid roboticsHumanoid robotics
Humanoid robotics
 
[26]자동화, 계륵에 살 붙이기 : Evolution of Android Automation Test
[26]자동화, 계륵에 살 붙이기 : Evolution of Android Automation Test[26]자동화, 계륵에 살 붙이기 : Evolution of Android Automation Test
[26]자동화, 계륵에 살 붙이기 : Evolution of Android Automation Test
 
Humanoid Service Robots: Challenges and Opportunities in Healthcare
Humanoid Service Robots: Challenges and Opportunities in HealthcareHumanoid Service Robots: Challenges and Opportunities in Healthcare
Humanoid Service Robots: Challenges and Opportunities in Healthcare
 
2015 py con word2vec이 추천시스템을 만났을때
2015 py con word2vec이 추천시스템을 만났을때 2015 py con word2vec이 추천시스템을 만났을때
2015 py con word2vec이 추천시스템을 만났을때
 
Intelligent mobile Robotics & Perception SystemsIntelligent mobile Robotics ...
Intelligent mobile Robotics  & Perception SystemsIntelligent mobile Robotics ...Intelligent mobile Robotics  & Perception SystemsIntelligent mobile Robotics ...
Intelligent mobile Robotics & Perception SystemsIntelligent mobile Robotics ...
 
Lecture 1 - Introduction to Robotics.pptx
Lecture 1 - Introduction to Robotics.pptxLecture 1 - Introduction to Robotics.pptx
Lecture 1 - Introduction to Robotics.pptx
 

En vedette

[141] 오픈소스를 쓰려는 자, 리베이스의 무게를 견뎌라
[141] 오픈소스를 쓰려는 자, 리베이스의 무게를 견뎌라[141] 오픈소스를 쓰려는 자, 리베이스의 무게를 견뎌라
[141] 오픈소스를 쓰려는 자, 리베이스의 무게를 견뎌라NAVER D2
 
밑바닥부터시작하는360뷰어
밑바닥부터시작하는360뷰어밑바닥부터시작하는360뷰어
밑바닥부터시작하는360뷰어NAVER D2
 
[131]chromium binging 기술을 node.js에 적용해보자
[131]chromium binging 기술을 node.js에 적용해보자[131]chromium binging 기술을 node.js에 적용해보자
[131]chromium binging 기술을 node.js에 적용해보자NAVER D2
 
[124]자율주행과 기계학습
[124]자율주행과 기계학습[124]자율주행과 기계학습
[124]자율주행과 기계학습NAVER D2
 
[141]네이버랩스의 로보틱스 연구 소개
[141]네이버랩스의 로보틱스 연구 소개[141]네이버랩스의 로보틱스 연구 소개
[141]네이버랩스의 로보틱스 연구 소개NAVER D2
 
[143]알파글래스의 개발과정으로 알아보는 ar 스마트글래스 광학 시스템
[143]알파글래스의 개발과정으로 알아보는 ar 스마트글래스 광학 시스템 [143]알파글래스의 개발과정으로 알아보는 ar 스마트글래스 광학 시스템
[143]알파글래스의 개발과정으로 알아보는 ar 스마트글래스 광학 시스템 NAVER D2
 
[132]웨일 브라우저 1년 그리고 미래
[132]웨일 브라우저 1년 그리고 미래[132]웨일 브라우저 1년 그리고 미래
[132]웨일 브라우저 1년 그리고 미래NAVER D2
 
[111]open, share, enjoy 네이버의 오픈소스 활동
[111]open, share, enjoy 네이버의 오픈소스 활동[111]open, share, enjoy 네이버의 오픈소스 활동
[111]open, share, enjoy 네이버의 오픈소스 활동NAVER D2
 
웨일 보안 이야기
웨일 보안 이야기웨일 보안 이야기
웨일 보안 이야기NAVER D2
 
[112]clova platform 인공지능을 엮는 기술
[112]clova platform 인공지능을 엮는 기술[112]clova platform 인공지능을 엮는 기술
[112]clova platform 인공지능을 엮는 기술NAVER D2
 
[123]동네 커피샵도 사이렌 오더를 쓸 수 있을까
[123]동네 커피샵도 사이렌 오더를 쓸 수 있을까[123]동네 커피샵도 사이렌 오더를 쓸 수 있을까
[123]동네 커피샵도 사이렌 오더를 쓸 수 있을까NAVER D2
 
what is_tabs_share
what is_tabs_sharewhat is_tabs_share
what is_tabs_shareNAVER D2
 
웨일브라우저 성능 및 메모리 최적화
웨일브라우저 성능 및 메모리 최적화웨일브라우저 성능 및 메모리 최적화
웨일브라우저 성능 및 메모리 최적화NAVER D2
 
[113]how can realm_make_efficient_mobile_database
[113]how can realm_make_efficient_mobile_database[113]how can realm_make_efficient_mobile_database
[113]how can realm_make_efficient_mobile_databaseNAVER D2
 
[135] 오픈소스 데이터베이스, 은행 서비스에 첫발을 내밀다.
[135] 오픈소스 데이터베이스, 은행 서비스에 첫발을 내밀다.[135] 오픈소스 데이터베이스, 은행 서비스에 첫발을 내밀다.
[135] 오픈소스 데이터베이스, 은행 서비스에 첫발을 내밀다.NAVER D2
 
[125] 머신러닝으로 쏟아지는 유저 cs 답변하기
[125] 머신러닝으로 쏟아지는 유저 cs 답변하기[125] 머신러닝으로 쏟아지는 유저 cs 답변하기
[125] 머신러닝으로 쏟아지는 유저 cs 답변하기NAVER D2
 
[115]14일만에 깃헙 스타 1,000개 받은 차트 오픈소스 개발기
[115]14일만에 깃헙 스타 1,000개 받은 차트 오픈소스 개발기[115]14일만에 깃헙 스타 1,000개 받은 차트 오픈소스 개발기
[115]14일만에 깃헙 스타 1,000개 받은 차트 오픈소스 개발기NAVER D2
 
[213]building ai to recreate our visual world
[213]building ai to recreate our visual world[213]building ai to recreate our visual world
[213]building ai to recreate our visual worldNAVER D2
 
[241]large scale search with polysemous codes
[241]large scale search with polysemous codes[241]large scale search with polysemous codes
[241]large scale search with polysemous codesNAVER D2
 
[222]neural machine translation (nmt) 동작의 시각화 및 분석 방법
[222]neural machine translation (nmt) 동작의 시각화 및 분석 방법[222]neural machine translation (nmt) 동작의 시각화 및 분석 방법
[222]neural machine translation (nmt) 동작의 시각화 및 분석 방법NAVER D2
 

En vedette (20)

[141] 오픈소스를 쓰려는 자, 리베이스의 무게를 견뎌라
[141] 오픈소스를 쓰려는 자, 리베이스의 무게를 견뎌라[141] 오픈소스를 쓰려는 자, 리베이스의 무게를 견뎌라
[141] 오픈소스를 쓰려는 자, 리베이스의 무게를 견뎌라
 
밑바닥부터시작하는360뷰어
밑바닥부터시작하는360뷰어밑바닥부터시작하는360뷰어
밑바닥부터시작하는360뷰어
 
[131]chromium binging 기술을 node.js에 적용해보자
[131]chromium binging 기술을 node.js에 적용해보자[131]chromium binging 기술을 node.js에 적용해보자
[131]chromium binging 기술을 node.js에 적용해보자
 
[124]자율주행과 기계학습
[124]자율주행과 기계학습[124]자율주행과 기계학습
[124]자율주행과 기계학습
 
[141]네이버랩스의 로보틱스 연구 소개
[141]네이버랩스의 로보틱스 연구 소개[141]네이버랩스의 로보틱스 연구 소개
[141]네이버랩스의 로보틱스 연구 소개
 
[143]알파글래스의 개발과정으로 알아보는 ar 스마트글래스 광학 시스템
[143]알파글래스의 개발과정으로 알아보는 ar 스마트글래스 광학 시스템 [143]알파글래스의 개발과정으로 알아보는 ar 스마트글래스 광학 시스템
[143]알파글래스의 개발과정으로 알아보는 ar 스마트글래스 광학 시스템
 
[132]웨일 브라우저 1년 그리고 미래
[132]웨일 브라우저 1년 그리고 미래[132]웨일 브라우저 1년 그리고 미래
[132]웨일 브라우저 1년 그리고 미래
 
[111]open, share, enjoy 네이버의 오픈소스 활동
[111]open, share, enjoy 네이버의 오픈소스 활동[111]open, share, enjoy 네이버의 오픈소스 활동
[111]open, share, enjoy 네이버의 오픈소스 활동
 
웨일 보안 이야기
웨일 보안 이야기웨일 보안 이야기
웨일 보안 이야기
 
[112]clova platform 인공지능을 엮는 기술
[112]clova platform 인공지능을 엮는 기술[112]clova platform 인공지능을 엮는 기술
[112]clova platform 인공지능을 엮는 기술
 
[123]동네 커피샵도 사이렌 오더를 쓸 수 있을까
[123]동네 커피샵도 사이렌 오더를 쓸 수 있을까[123]동네 커피샵도 사이렌 오더를 쓸 수 있을까
[123]동네 커피샵도 사이렌 오더를 쓸 수 있을까
 
what is_tabs_share
what is_tabs_sharewhat is_tabs_share
what is_tabs_share
 
웨일브라우저 성능 및 메모리 최적화
웨일브라우저 성능 및 메모리 최적화웨일브라우저 성능 및 메모리 최적화
웨일브라우저 성능 및 메모리 최적화
 
[113]how can realm_make_efficient_mobile_database
[113]how can realm_make_efficient_mobile_database[113]how can realm_make_efficient_mobile_database
[113]how can realm_make_efficient_mobile_database
 
[135] 오픈소스 데이터베이스, 은행 서비스에 첫발을 내밀다.
[135] 오픈소스 데이터베이스, 은행 서비스에 첫발을 내밀다.[135] 오픈소스 데이터베이스, 은행 서비스에 첫발을 내밀다.
[135] 오픈소스 데이터베이스, 은행 서비스에 첫발을 내밀다.
 
[125] 머신러닝으로 쏟아지는 유저 cs 답변하기
[125] 머신러닝으로 쏟아지는 유저 cs 답변하기[125] 머신러닝으로 쏟아지는 유저 cs 답변하기
[125] 머신러닝으로 쏟아지는 유저 cs 답변하기
 
[115]14일만에 깃헙 스타 1,000개 받은 차트 오픈소스 개발기
[115]14일만에 깃헙 스타 1,000개 받은 차트 오픈소스 개발기[115]14일만에 깃헙 스타 1,000개 받은 차트 오픈소스 개발기
[115]14일만에 깃헙 스타 1,000개 받은 차트 오픈소스 개발기
 
[213]building ai to recreate our visual world
[213]building ai to recreate our visual world[213]building ai to recreate our visual world
[213]building ai to recreate our visual world
 
[241]large scale search with polysemous codes
[241]large scale search with polysemous codes[241]large scale search with polysemous codes
[241]large scale search with polysemous codes
 
[222]neural machine translation (nmt) 동작의 시각화 및 분석 방법
[222]neural machine translation (nmt) 동작의 시각화 및 분석 방법[222]neural machine translation (nmt) 동작의 시각화 및 분석 방법
[222]neural machine translation (nmt) 동작의 시각화 및 분석 방법
 

Similaire à [142] 생체 이해에 기반한 로봇 – 고성능 로봇에게 인간의 유연함과 안전성 부여하기

MARK ROBOTIC ARM.ppt
MARK ROBOTIC ARM.pptMARK ROBOTIC ARM.ppt
MARK ROBOTIC ARM.ppttffttfyyf
 
MARK ROBOTIC ARM.ppt
MARK ROBOTIC ARM.pptMARK ROBOTIC ARM.ppt
MARK ROBOTIC ARM.pptAfstddrrdv
 
Industrial robot
Industrial robotIndustrial robot
Industrial robotJatin Garg
 
Review on Development of Industrial Robotic Arm
Review on Development of Industrial Robotic ArmReview on Development of Industrial Robotic Arm
Review on Development of Industrial Robotic ArmIRJET Journal
 
Pick & place robot ppt
Pick & place robot pptPick & place robot ppt
Pick & place robot pptRahul Banerjee
 
Continuum Robots and Tactile Sensors
Continuum Robots and Tactile SensorsContinuum Robots and Tactile Sensors
Continuum Robots and Tactile SensorsTacuna Systems
 
Space robotics
Space roboticsSpace robotics
Space roboticsMU
 
Vehicle Controlled by Hand Gesture Using Raspberry pi
Vehicle Controlled by Hand Gesture Using Raspberry piVehicle Controlled by Hand Gesture Using Raspberry pi
Vehicle Controlled by Hand Gesture Using Raspberry piIRJET Journal
 
Computer aided manufacturing robotic systems
Computer aided manufacturing robotic systemsComputer aided manufacturing robotic systems
Computer aided manufacturing robotic systemsPrasanth Kumar RAGUPATHY
 
DEVENDRA SHRIVASH ROBOTICS .pptx
DEVENDRA SHRIVASH ROBOTICS .pptxDEVENDRA SHRIVASH ROBOTICS .pptx
DEVENDRA SHRIVASH ROBOTICS .pptxDEVENDRA SHRIVASH
 
Geasture Control Robotic Arm
Geasture Control Robotic ArmGeasture Control Robotic Arm
Geasture Control Robotic ArmSree Harsha
 
Wireless bomb disposal robot ppt
Wireless bomb disposal robot pptWireless bomb disposal robot ppt
Wireless bomb disposal robot pptAbhishek Gupta
 
Robotics Group 10 (Control Schemes) cse.pdf
Robotics Group 10  (Control Schemes) cse.pdfRobotics Group 10  (Control Schemes) cse.pdf
Robotics Group 10 (Control Schemes) cse.pdfsahilsajad201
 

Similaire à [142] 생체 이해에 기반한 로봇 – 고성능 로봇에게 인간의 유연함과 안전성 부여하기 (20)

MARK ROBOTIC ARM.ppt
MARK ROBOTIC ARM.pptMARK ROBOTIC ARM.ppt
MARK ROBOTIC ARM.ppt
 
MARK ROBOTIC ARM.ppt
MARK ROBOTIC ARM.pptMARK ROBOTIC ARM.ppt
MARK ROBOTIC ARM.ppt
 
Robot arm ppt
Robot arm pptRobot arm ppt
Robot arm ppt
 
robotic arm
robotic arm robotic arm
robotic arm
 
Industrial robot
Industrial robotIndustrial robot
Industrial robot
 
Review on Development of Industrial Robotic Arm
Review on Development of Industrial Robotic ArmReview on Development of Industrial Robotic Arm
Review on Development of Industrial Robotic Arm
 
Pick and place
Pick and placePick and place
Pick and place
 
pick and place ppt.ppt
pick and place ppt.pptpick and place ppt.ppt
pick and place ppt.ppt
 
Pick & place robot ppt
Pick & place robot pptPick & place robot ppt
Pick & place robot ppt
 
Continuum Robots and Tactile Sensors
Continuum Robots and Tactile SensorsContinuum Robots and Tactile Sensors
Continuum Robots and Tactile Sensors
 
Space robotics
Space roboticsSpace robotics
Space robotics
 
Robotic 6DOF ARM
Robotic 6DOF ARMRobotic 6DOF ARM
Robotic 6DOF ARM
 
Vehicle Controlled by Hand Gesture Using Raspberry pi
Vehicle Controlled by Hand Gesture Using Raspberry piVehicle Controlled by Hand Gesture Using Raspberry pi
Vehicle Controlled by Hand Gesture Using Raspberry pi
 
Computer aided manufacturing robotic systems
Computer aided manufacturing robotic systemsComputer aided manufacturing robotic systems
Computer aided manufacturing robotic systems
 
Robotics
RoboticsRobotics
Robotics
 
Cobotics
Cobotics Cobotics
Cobotics
 
DEVENDRA SHRIVASH ROBOTICS .pptx
DEVENDRA SHRIVASH ROBOTICS .pptxDEVENDRA SHRIVASH ROBOTICS .pptx
DEVENDRA SHRIVASH ROBOTICS .pptx
 
Geasture Control Robotic Arm
Geasture Control Robotic ArmGeasture Control Robotic Arm
Geasture Control Robotic Arm
 
Wireless bomb disposal robot ppt
Wireless bomb disposal robot pptWireless bomb disposal robot ppt
Wireless bomb disposal robot ppt
 
Robotics Group 10 (Control Schemes) cse.pdf
Robotics Group 10  (Control Schemes) cse.pdfRobotics Group 10  (Control Schemes) cse.pdf
Robotics Group 10 (Control Schemes) cse.pdf
 

Plus de NAVER D2

[211] 인공지능이 인공지능 챗봇을 만든다
[211] 인공지능이 인공지능 챗봇을 만든다[211] 인공지능이 인공지능 챗봇을 만든다
[211] 인공지능이 인공지능 챗봇을 만든다NAVER D2
 
[233] 대형 컨테이너 클러스터에서의 고가용성 Network Load Balancing: Maglev Hashing Scheduler i...
[233] 대형 컨테이너 클러스터에서의 고가용성 Network Load Balancing: Maglev Hashing Scheduler i...[233] 대형 컨테이너 클러스터에서의 고가용성 Network Load Balancing: Maglev Hashing Scheduler i...
[233] 대형 컨테이너 클러스터에서의 고가용성 Network Load Balancing: Maglev Hashing Scheduler i...NAVER D2
 
[215] Druid로 쉽고 빠르게 데이터 분석하기
[215] Druid로 쉽고 빠르게 데이터 분석하기[215] Druid로 쉽고 빠르게 데이터 분석하기
[215] Druid로 쉽고 빠르게 데이터 분석하기NAVER D2
 
[245]Papago Internals: 모델분석과 응용기술 개발
[245]Papago Internals: 모델분석과 응용기술 개발[245]Papago Internals: 모델분석과 응용기술 개발
[245]Papago Internals: 모델분석과 응용기술 개발NAVER D2
 
[236] 스트림 저장소 최적화 이야기: 아파치 드루이드로부터 얻은 교훈
[236] 스트림 저장소 최적화 이야기: 아파치 드루이드로부터 얻은 교훈[236] 스트림 저장소 최적화 이야기: 아파치 드루이드로부터 얻은 교훈
[236] 스트림 저장소 최적화 이야기: 아파치 드루이드로부터 얻은 교훈NAVER D2
 
[235]Wikipedia-scale Q&A
[235]Wikipedia-scale Q&A[235]Wikipedia-scale Q&A
[235]Wikipedia-scale Q&ANAVER D2
 
[244]로봇이 현실 세계에 대해 학습하도록 만들기
[244]로봇이 현실 세계에 대해 학습하도록 만들기[244]로봇이 현실 세계에 대해 학습하도록 만들기
[244]로봇이 현실 세계에 대해 학습하도록 만들기NAVER D2
 
[243] Deep Learning to help student’s Deep Learning
[243] Deep Learning to help student’s Deep Learning[243] Deep Learning to help student’s Deep Learning
[243] Deep Learning to help student’s Deep LearningNAVER D2
 
[234]Fast & Accurate Data Annotation Pipeline for AI applications
[234]Fast & Accurate Data Annotation Pipeline for AI applications[234]Fast & Accurate Data Annotation Pipeline for AI applications
[234]Fast & Accurate Data Annotation Pipeline for AI applicationsNAVER D2
 
Old version: [233]대형 컨테이너 클러스터에서의 고가용성 Network Load Balancing
Old version: [233]대형 컨테이너 클러스터에서의 고가용성 Network Load BalancingOld version: [233]대형 컨테이너 클러스터에서의 고가용성 Network Load Balancing
Old version: [233]대형 컨테이너 클러스터에서의 고가용성 Network Load BalancingNAVER D2
 
[226]NAVER 광고 deep click prediction: 모델링부터 서빙까지
[226]NAVER 광고 deep click prediction: 모델링부터 서빙까지[226]NAVER 광고 deep click prediction: 모델링부터 서빙까지
[226]NAVER 광고 deep click prediction: 모델링부터 서빙까지NAVER D2
 
[225]NSML: 머신러닝 플랫폼 서비스하기 & 모델 튜닝 자동화하기
[225]NSML: 머신러닝 플랫폼 서비스하기 & 모델 튜닝 자동화하기[225]NSML: 머신러닝 플랫폼 서비스하기 & 모델 튜닝 자동화하기
[225]NSML: 머신러닝 플랫폼 서비스하기 & 모델 튜닝 자동화하기NAVER D2
 
[224]네이버 검색과 개인화
[224]네이버 검색과 개인화[224]네이버 검색과 개인화
[224]네이버 검색과 개인화NAVER D2
 
[216]Search Reliability Engineering (부제: 지진에도 흔들리지 않는 네이버 검색시스템)
[216]Search Reliability Engineering (부제: 지진에도 흔들리지 않는 네이버 검색시스템)[216]Search Reliability Engineering (부제: 지진에도 흔들리지 않는 네이버 검색시스템)
[216]Search Reliability Engineering (부제: 지진에도 흔들리지 않는 네이버 검색시스템)NAVER D2
 
[214] Ai Serving Platform: 하루 수 억 건의 인퍼런스를 처리하기 위한 고군분투기
[214] Ai Serving Platform: 하루 수 억 건의 인퍼런스를 처리하기 위한 고군분투기[214] Ai Serving Platform: 하루 수 억 건의 인퍼런스를 처리하기 위한 고군분투기
[214] Ai Serving Platform: 하루 수 억 건의 인퍼런스를 처리하기 위한 고군분투기NAVER D2
 
[213] Fashion Visual Search
[213] Fashion Visual Search[213] Fashion Visual Search
[213] Fashion Visual SearchNAVER D2
 
[232] TensorRT를 활용한 딥러닝 Inference 최적화
[232] TensorRT를 활용한 딥러닝 Inference 최적화[232] TensorRT를 활용한 딥러닝 Inference 최적화
[232] TensorRT를 활용한 딥러닝 Inference 최적화NAVER D2
 
[242]컴퓨터 비전을 이용한 실내 지도 자동 업데이트 방법: 딥러닝을 통한 POI 변화 탐지
[242]컴퓨터 비전을 이용한 실내 지도 자동 업데이트 방법: 딥러닝을 통한 POI 변화 탐지[242]컴퓨터 비전을 이용한 실내 지도 자동 업데이트 방법: 딥러닝을 통한 POI 변화 탐지
[242]컴퓨터 비전을 이용한 실내 지도 자동 업데이트 방법: 딥러닝을 통한 POI 변화 탐지NAVER D2
 
[212]C3, 데이터 처리에서 서빙까지 가능한 하둡 클러스터
[212]C3, 데이터 처리에서 서빙까지 가능한 하둡 클러스터[212]C3, 데이터 처리에서 서빙까지 가능한 하둡 클러스터
[212]C3, 데이터 처리에서 서빙까지 가능한 하둡 클러스터NAVER D2
 
[223]기계독해 QA: 검색인가, NLP인가?
[223]기계독해 QA: 검색인가, NLP인가?[223]기계독해 QA: 검색인가, NLP인가?
[223]기계독해 QA: 검색인가, NLP인가?NAVER D2
 

Plus de NAVER D2 (20)

[211] 인공지능이 인공지능 챗봇을 만든다
[211] 인공지능이 인공지능 챗봇을 만든다[211] 인공지능이 인공지능 챗봇을 만든다
[211] 인공지능이 인공지능 챗봇을 만든다
 
[233] 대형 컨테이너 클러스터에서의 고가용성 Network Load Balancing: Maglev Hashing Scheduler i...
[233] 대형 컨테이너 클러스터에서의 고가용성 Network Load Balancing: Maglev Hashing Scheduler i...[233] 대형 컨테이너 클러스터에서의 고가용성 Network Load Balancing: Maglev Hashing Scheduler i...
[233] 대형 컨테이너 클러스터에서의 고가용성 Network Load Balancing: Maglev Hashing Scheduler i...
 
[215] Druid로 쉽고 빠르게 데이터 분석하기
[215] Druid로 쉽고 빠르게 데이터 분석하기[215] Druid로 쉽고 빠르게 데이터 분석하기
[215] Druid로 쉽고 빠르게 데이터 분석하기
 
[245]Papago Internals: 모델분석과 응용기술 개발
[245]Papago Internals: 모델분석과 응용기술 개발[245]Papago Internals: 모델분석과 응용기술 개발
[245]Papago Internals: 모델분석과 응용기술 개발
 
[236] 스트림 저장소 최적화 이야기: 아파치 드루이드로부터 얻은 교훈
[236] 스트림 저장소 최적화 이야기: 아파치 드루이드로부터 얻은 교훈[236] 스트림 저장소 최적화 이야기: 아파치 드루이드로부터 얻은 교훈
[236] 스트림 저장소 최적화 이야기: 아파치 드루이드로부터 얻은 교훈
 
[235]Wikipedia-scale Q&A
[235]Wikipedia-scale Q&A[235]Wikipedia-scale Q&A
[235]Wikipedia-scale Q&A
 
[244]로봇이 현실 세계에 대해 학습하도록 만들기
[244]로봇이 현실 세계에 대해 학습하도록 만들기[244]로봇이 현실 세계에 대해 학습하도록 만들기
[244]로봇이 현실 세계에 대해 학습하도록 만들기
 
[243] Deep Learning to help student’s Deep Learning
[243] Deep Learning to help student’s Deep Learning[243] Deep Learning to help student’s Deep Learning
[243] Deep Learning to help student’s Deep Learning
 
[234]Fast & Accurate Data Annotation Pipeline for AI applications
[234]Fast & Accurate Data Annotation Pipeline for AI applications[234]Fast & Accurate Data Annotation Pipeline for AI applications
[234]Fast & Accurate Data Annotation Pipeline for AI applications
 
Old version: [233]대형 컨테이너 클러스터에서의 고가용성 Network Load Balancing
Old version: [233]대형 컨테이너 클러스터에서의 고가용성 Network Load BalancingOld version: [233]대형 컨테이너 클러스터에서의 고가용성 Network Load Balancing
Old version: [233]대형 컨테이너 클러스터에서의 고가용성 Network Load Balancing
 
[226]NAVER 광고 deep click prediction: 모델링부터 서빙까지
[226]NAVER 광고 deep click prediction: 모델링부터 서빙까지[226]NAVER 광고 deep click prediction: 모델링부터 서빙까지
[226]NAVER 광고 deep click prediction: 모델링부터 서빙까지
 
[225]NSML: 머신러닝 플랫폼 서비스하기 & 모델 튜닝 자동화하기
[225]NSML: 머신러닝 플랫폼 서비스하기 & 모델 튜닝 자동화하기[225]NSML: 머신러닝 플랫폼 서비스하기 & 모델 튜닝 자동화하기
[225]NSML: 머신러닝 플랫폼 서비스하기 & 모델 튜닝 자동화하기
 
[224]네이버 검색과 개인화
[224]네이버 검색과 개인화[224]네이버 검색과 개인화
[224]네이버 검색과 개인화
 
[216]Search Reliability Engineering (부제: 지진에도 흔들리지 않는 네이버 검색시스템)
[216]Search Reliability Engineering (부제: 지진에도 흔들리지 않는 네이버 검색시스템)[216]Search Reliability Engineering (부제: 지진에도 흔들리지 않는 네이버 검색시스템)
[216]Search Reliability Engineering (부제: 지진에도 흔들리지 않는 네이버 검색시스템)
 
[214] Ai Serving Platform: 하루 수 억 건의 인퍼런스를 처리하기 위한 고군분투기
[214] Ai Serving Platform: 하루 수 억 건의 인퍼런스를 처리하기 위한 고군분투기[214] Ai Serving Platform: 하루 수 억 건의 인퍼런스를 처리하기 위한 고군분투기
[214] Ai Serving Platform: 하루 수 억 건의 인퍼런스를 처리하기 위한 고군분투기
 
[213] Fashion Visual Search
[213] Fashion Visual Search[213] Fashion Visual Search
[213] Fashion Visual Search
 
[232] TensorRT를 활용한 딥러닝 Inference 최적화
[232] TensorRT를 활용한 딥러닝 Inference 최적화[232] TensorRT를 활용한 딥러닝 Inference 최적화
[232] TensorRT를 활용한 딥러닝 Inference 최적화
 
[242]컴퓨터 비전을 이용한 실내 지도 자동 업데이트 방법: 딥러닝을 통한 POI 변화 탐지
[242]컴퓨터 비전을 이용한 실내 지도 자동 업데이트 방법: 딥러닝을 통한 POI 변화 탐지[242]컴퓨터 비전을 이용한 실내 지도 자동 업데이트 방법: 딥러닝을 통한 POI 변화 탐지
[242]컴퓨터 비전을 이용한 실내 지도 자동 업데이트 방법: 딥러닝을 통한 POI 변화 탐지
 
[212]C3, 데이터 처리에서 서빙까지 가능한 하둡 클러스터
[212]C3, 데이터 처리에서 서빙까지 가능한 하둡 클러스터[212]C3, 데이터 처리에서 서빙까지 가능한 하둡 클러스터
[212]C3, 데이터 처리에서 서빙까지 가능한 하둡 클러스터
 
[223]기계독해 QA: 검색인가, NLP인가?
[223]기계독해 QA: 검색인가, NLP인가?[223]기계독해 QA: 검색인가, NLP인가?
[223]기계독해 QA: 검색인가, NLP인가?
 

Dernier

How to write a Business Continuity Plan
How to write a Business Continuity PlanHow to write a Business Continuity Plan
How to write a Business Continuity PlanDatabarracks
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024BookNet Canada
 
What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024Stephanie Beckett
 
Developer Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLDeveloper Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLScyllaDB
 
How AI, OpenAI, and ChatGPT impact business and software.
How AI, OpenAI, and ChatGPT impact business and software.How AI, OpenAI, and ChatGPT impact business and software.
How AI, OpenAI, and ChatGPT impact business and software.Curtis Poe
 
Passkey Providers and Enabling Portability: FIDO Paris Seminar.pptx
Passkey Providers and Enabling Portability: FIDO Paris Seminar.pptxPasskey Providers and Enabling Portability: FIDO Paris Seminar.pptx
Passkey Providers and Enabling Portability: FIDO Paris Seminar.pptxLoriGlavin3
 
A Journey Into the Emotions of Software Developers
A Journey Into the Emotions of Software DevelopersA Journey Into the Emotions of Software Developers
A Journey Into the Emotions of Software DevelopersNicole Novielli
 
A Deep Dive on Passkeys: FIDO Paris Seminar.pptx
A Deep Dive on Passkeys: FIDO Paris Seminar.pptxA Deep Dive on Passkeys: FIDO Paris Seminar.pptx
A Deep Dive on Passkeys: FIDO Paris Seminar.pptxLoriGlavin3
 
Unraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfUnraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfAlex Barbosa Coqueiro
 
Artificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxArtificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxhariprasad279825
 
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptxMerck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptxLoriGlavin3
 
TeamStation AI System Report LATAM IT Salaries 2024
TeamStation AI System Report LATAM IT Salaries 2024TeamStation AI System Report LATAM IT Salaries 2024
TeamStation AI System Report LATAM IT Salaries 2024Lonnie McRorey
 
WordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your BrandWordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your Brandgvaughan
 
From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .Alan Dix
 
unit 4 immunoblotting technique complete.pptx
unit 4 immunoblotting technique complete.pptxunit 4 immunoblotting technique complete.pptx
unit 4 immunoblotting technique complete.pptxBkGupta21
 
"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii SoldatenkoFwdays
 
Rise of the Machines: Known As Drones...
Rise of the Machines: Known As Drones...Rise of the Machines: Known As Drones...
Rise of the Machines: Known As Drones...Rick Flair
 
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Mark Simos
 
DevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache MavenDevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache MavenHervé Boutemy
 
SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024Lorenzo Miniero
 

Dernier (20)

How to write a Business Continuity Plan
How to write a Business Continuity PlanHow to write a Business Continuity Plan
How to write a Business Continuity Plan
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
 
What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024
 
Developer Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLDeveloper Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQL
 
How AI, OpenAI, and ChatGPT impact business and software.
How AI, OpenAI, and ChatGPT impact business and software.How AI, OpenAI, and ChatGPT impact business and software.
How AI, OpenAI, and ChatGPT impact business and software.
 
Passkey Providers and Enabling Portability: FIDO Paris Seminar.pptx
Passkey Providers and Enabling Portability: FIDO Paris Seminar.pptxPasskey Providers and Enabling Portability: FIDO Paris Seminar.pptx
Passkey Providers and Enabling Portability: FIDO Paris Seminar.pptx
 
A Journey Into the Emotions of Software Developers
A Journey Into the Emotions of Software DevelopersA Journey Into the Emotions of Software Developers
A Journey Into the Emotions of Software Developers
 
A Deep Dive on Passkeys: FIDO Paris Seminar.pptx
A Deep Dive on Passkeys: FIDO Paris Seminar.pptxA Deep Dive on Passkeys: FIDO Paris Seminar.pptx
A Deep Dive on Passkeys: FIDO Paris Seminar.pptx
 
Unraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfUnraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdf
 
Artificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxArtificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptx
 
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptxMerck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptx
 
TeamStation AI System Report LATAM IT Salaries 2024
TeamStation AI System Report LATAM IT Salaries 2024TeamStation AI System Report LATAM IT Salaries 2024
TeamStation AI System Report LATAM IT Salaries 2024
 
WordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your BrandWordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your Brand
 
From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .
 
unit 4 immunoblotting technique complete.pptx
unit 4 immunoblotting technique complete.pptxunit 4 immunoblotting technique complete.pptx
unit 4 immunoblotting technique complete.pptx
 
"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko
 
Rise of the Machines: Known As Drones...
Rise of the Machines: Known As Drones...Rise of the Machines: Known As Drones...
Rise of the Machines: Known As Drones...
 
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
 
DevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache MavenDevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache Maven
 
SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024
 

[142] 생체 이해에 기반한 로봇 – 고성능 로봇에게 인간의 유연함과 안전성 부여하기

  • 1. 생체 이해에 기반한 로봇 김용재 Ph.D. IRIM Lab. Koreatech 고성능 로봇에게 인간의 유연함과 안정성 부여하기
  • 2. PROFESSIONAL EXPERINECE 2014.3 - Present Assistant Professor, KoreaTech 2012.3 - 2014.3 Research Staff Member, Intelligent Robot Group, SAIT 2011. 3 - 2012.2 Principal Engineer and Visiting Engineer of MIT Manufacturing Tech. Center, Samsung Electronics 2003. 3 - 2011.2 Senior Engineer, Robot Technology Team Manufacturing Tech. Center, Samsung Electronics Education : ~2003, Ph.D in KAIST ~1998, M.S in KAIST ~1996, B.S in KAIST Major: Electrical Engineering and Computer Science Yong Jae Kim, Ph.D. RESEARCH INTERESTS - Mechanism design and Control of Wearable Robots, Surgical Robots, Humanoid & Interactive Mobile Robots - Design and Analysis of Flexible Robots and Backdrivable Mechanisms - Compliance Control for Backdrivable Mechanism & Redundant manipulators - Idea Generation of Novel Robot System (22 US patents, TRIZ expert Level II)
  • 3. ※ Ref: ABB Yumi Personal Robot Public Robot “iMaro-PSR” Humanoid “RoboRay” Single Port Surgical Robot Samsung Cleaning robot 2004 2005 2006 20102007 2009 2012 2013 Samsung Robot Commercialization TF Cobot “Rayarm” Walking Assist Robot SAMSUNG ELECTRONICS (2003-2013)
  • 4. Flexible Human Assist Robot (Confidential Project) (2014) Compliant Gripper (미래부) •2016~ Compression support frame Kevlar Belt Ni-Ti Cable Non-stretchable Fabric Conduit Power Assist Glove •2016~2004 2005 2006 2015 2016 2017 High-Payload Gravity Compensation Manipulator (Confidential Project) Torque sensing Manipulator (Confidential Project) •2015~2004 2005 2006 20152014 2016 DARPA Robotics Challenge (산자부) (2014-15) LIMS: Highly Interactive Manipulator (2014 ~) Anthropomorphic dual arm robot for High speed Interaction (LAVER LABS) •2009 •2012 •2013 •2010 2012 2014 2015 2016 20172010 Interactive Robotics & Innovative Mechanism LabIRIM •2014 KOREATECH(2014-2017)
  • 5. TABLE OF CONTENTS 1_ 인간의 몸과 로봇 메커니즘 “산업용 로봇은 왜 박수를 칠 수 없을까?” 2_ 생체 이해에 기반한 로봇 핸드 “유연성과 정밀성을 모두 갖춘 로봇이 가능할까?” 2.1_ RoboRay Hand: 고자유도 휴머노이드 핸드 2.2_ TATH Glove : 파지력 증강 소프트 글러브 2.3_ Soft Parallel Gripper: 정밀 유연 그리퍼 3_ 생체 이해에 기반한 로봇 팔 “하이 파이브가 가능한 로봇의 조건은?” 3.1_중력보상 로봇, 가변강성 로봇 3.2_ LIMS2–AMBIDEX: 경량 고강성 로봇 팔 4_ 인간과 협력하는 로봇
  • 6. 1.인간의 몸과 로봇 메커니즘 “산업용 로봇은 왜 박수를 칠 수 없을까?”
  • 7. Robot Manipulators vs. Human Arms Weight : Joint speed : TCP speed : Payload : Rotational inertia : Stiffness : Repeatability : Control frequency : 23.9 kg 90 ~180 deg/sec Approx. 1.5m/sec 7 kg over 5kgm2 10,000~40,000 Nm/rad 0.1 mm 1kHz ( 3~4kHz, joint control) Weight : Joint speed : TCP speed : Payload : Rotational inertia : Stiffness : Repeatability : Control frequency : 3.9 kg 400 ~ 1200 deg/sec over 15m/sec - 0.49kgm2 350 Nm/rad (elbow) ? 10Hz (40Hz, uncond. reflex) Video Ref: ABB Robotics - Fanta Can Challenge- Level II - Superior Motion Control https://www.youtube.com/watch?v=SOESSCXGhFo Video Ref: Last 5 Minutes of Super Bowl 51 https://www.youtube.com/watch?v=iN8Rfwr6L2U Pull Up & Chin Up Progression Guide incl. 10+ Exercises https://m.youtube.com/watch?v=CdtrfXK7bcg Men's Archery Individual Gold Medal Match | Rio 2016 Replay https://www.youtube.com/watch?v=rzj4FFi7wt8
  • 8. Robot Grippers vs. Human Hands Weight : Degrees of Freedom : Joint speed : Fingertip force : Grasp force : No. of Muscles : 1.5 kg 23 DOF Over 500 deg/sec 43 N (index, straight pose) 370 N over 30 muscles Video Ref.: Grasping different objects quickly https://www.youtube.com/watch?v=ZNjbyCWt93k Flexible Robot Gripper: 2-finger Adaptive Electric Robot Gripper by Robotiq https://www.youtube.com/watch?v=nkGuI4uiSLM Soft Robotics Inc. - Adaptive Grasping of Common E-Commerce Items https://www.youtube.com/watch?v=vRQZ5bqwDbg Video Ref.: Barbell Deadlift Bent Row Complex https://www.youtube.com/watch?v=PUNxkzCjWNk World's most complicated watch https://www.youtube.com/watch?v=I1L15xehfEA Slow Motion Rubik's Cube Solve https://www.youtube.com/watch?v=gEPgcWx1JMs
  • 9. Contact materials? Sensors? Control Algorithm? … Robot Grippers vs. Human Hands Actuation - More than 30 muscles - Highly redundant! - Tendon driven Actuation - 1~ 16 motors - Difficult to increase DOF. - Gear or wire driven 인체는 구동기의 증가에 대한 부담 이 없다!! Lubrication - Lub. by Synovial fluid - Friction coeff.: 0.003~0.01 Lubrication - Bearing Friction coeff.: approx. 0.026 - Teflon : 0.04 베어링 대비 2.6~8.6 분의 1 저마찰 윤활유가 평생 공급된다!! Contact materials? Sensors? Control Algorithm? … 생체 모양만 모사하는 것은 무의미하다! 어느 level에서 모사해야 할까? Joints - Rolling & sliding - Aligned by ligaments Joints - Revolute joints - Bearing for low frictions - Vulnerable to Impact Maintenance Free!! 평생 장력이 유지 되는 인대구조.
  • 10. 인간의 능력을 모사하는 로봇들 Precise, High Payload High Speed Soft/flexible, Interactive, Compliant 유연한 로봇은 정밀한 로봇이 될 수 없는가? 인간은 어디에 위치하는가? Industrial Robot + force sensors + contact detection + … Soft Robot + Increase payload + Position sensors + …
  • 11. 2. 생체 이해에 기반한 로봇 핸드 “유연성과 정밀성을 모두 갖춘 로봇이 가능할까?”
  • 12. C6M Shadow Robotics 3-Finger Gripping Hand Schunk 5-Finger Hand Schunk i-Limb Ultra Touch Bionics Robonaut hand NASA Barrett hand Barrett Corp. Human Hand 27 Bones, 23 DOF, Over 30 muscles Industrial Prosthetic Research & General purpose C6M DLR RoboRay hand Samsung Humanoid The goal : Precise Manipulation + Compliant Power Grasping 인간과 유사한 크기/모양/무게, 마찰 최소화 및 역구동성 최대화 RoboRay Hand ( Samsung Project, published in ICRA2014 )
  • 13. RoboRay Hand : Highly backdrivable Life-Size Robotic Hand – 1.59kg, 12 DOF for fingers and 2 DOF for the wrist – Fingertip force 15N / Contact force detection 0.196N – Tendon-driven mechanism actuated by ball screws and BLDC motors Weight 1.59kg (including wrist and forearm) Dimensions Hand 160 x 80 x 45 mm Forearm 186 x 76 x 83 mm DOF Fingers 12 DOF / 5 Fingers Wrist 2 DOF Payload Peak Fingertip Force 15N (Stretched Pose) Speed/ Reduction Ratio MPR joint 800deg/sec, 47:1 MPP joint 700deg/sec, 57:1 PIP joint 450deg/sec, 82:1 Sensing Minimum Force Detection with compensation : 0.196N w.o. compensation : 0.735N Actuation Tendon driven, actuated by ball screws and BLDC Motors Electronics DC 12V, DSP TMS320F2812 EtherCAT Communication 2.1 RoboRay Hand- Specification ( Samsung Project, published in ICRA2014 )
  • 14. RoboRay Hand - Samsung Project - Published in ICRA2014 “RoboRay Hand : A Highly Backdrivable Robotic Hand with Sensorless Contact Force Measurements” Yong-Jae Kim, et al, ICRA2014
  • 15. Orangutan Baboon Human RoboRay 40 57-58 60 App. 100 Opposability among primates Metacarpus of the thumb 1. Biomimetic Design based on functionality of human hands 2.1 Mechanical Design - Configuration High Opposability – Human thumb is highly opposable due to mobile metacarpus Finger Design Considering Opposability
  • 16. 2. Actuator Configuration Based on the Grasping Functions Precise Manipulation Grasp motor fixed Pose motor in motion - Fixing the grasp motor and actuating pose motor enable precise adjustment. Compliant Grasping Pose motor off Grasp motor In motion - Turning the pose motor off and actuating the grasp motor make compliant grasping High force grasp motor in the forearm Small Pose Motor in the Palm Proposed Configuration MP joint PIP joint DIP joint Underactuated Hands - Inherent mechanical compliance - Difficult to manipulate object precisely Dexterous Hands - Fully controllable, Precise manipulation - Subject to be bulky, or slow and weak 2.1 Mechanical Design – Finger Actuation Concept
  • 17. 3. Tension Decoupling Wrist Mechanism using Rolling Joint Transmission using Conduits - Saving space, High force - Frictional, Limited Range of Motion Built-in Actuators in the Palm - Relatively Simple - Subject to be bulky, weak or slow Proposed Wrist Mechanism Wrist Pitch Motion Wrist Roll Motion Pitch Direction Roll Direction 2p 1p r Rolling Joint for Wrist Pitch Offset Pivot for Wrist Roll Composed 2-DOF Joint Rolling Joint 2.1 Mechanical Design - Wrist Decoupling Concept
  • 18. Stand-alone Performance Test Gestrure and Grasping Test (Prototype of RoboRay hand) Gesture and Manipulation Test (RoboRay hand without Covers) Performance back-drivability - backdrivable force : 0.735N - Mechanical Efficiency : App. 89.5% - Stiffness : App. 1.7N/mm Bearing Efficiency 97.0% No. Radius (mm) Angle (deg) Efficiency (%) 1 3.2 90 99.4 2 4.0 180 99.4 3 10.0 180 99.7 4 10.0 180 99.7 5 4.0 90 99.5 6 4.0 180 99.4 7 10.0 90 99.8 Ball Screw Efficiency 95% Actuator Bushing Efficiency 95% Total Efficiency 89.4% 2.1 Mechanical Design - Wrist Decoupling Concept
  • 19. Flexible/Soft and Safe Rigid, Precise and Strong Human hand - 유연성 + 정밀성 - 기능 수준의 생체 모사 - 기존의 기계적인 요소품 사용 2.1 RoboRay Hand
  • 20. 2.1 Flexible/Soft but Strong? Flexible/Soft and Safe Rigid, Precise and Strong Human hand Flexible/Soft + High payload ? - 유연한 구조인 동시에 강한 힘을 전달할 수 있다면...
  • 21. FDC Structure (Force Distributing Compliant Structure) - 말단부의 강성을 높이거나 조절 가능 - 구조에 따라 규제되는 자유도를 선택가능 Vertical & Lateral Motion Restricted torsional motion permitted Fin-Gripper (FESTO. Co) Vertical Motion Restricted Lateral & torsional motion permitted Compliant to side force Vertical & torsional Motion Restricted Lateral motion permitted It can withstand Vertical & torsional Loads! 기존 구조물 Cantilever Beam - Stiff proximal part - Soft distal part High Stiffness Low Stiffness 2.2 New Opportunities by FDC structure It can withstand Vertical, lateral & torsional Loads! All 3 motions restricted 단순 Cantilever Beam
  • 22. Flexible beam Flexible beam Dovetails for prismatic motion Walking Assist Robot, GEMS Younbaek lee, IROS2017 SAIT, Samsung electronics GEMS (SAIT, Samsung) Small deformationF Large deflection F Fin-gripper, Festo. Fin-gripper (FESTO Co.) 2.2 New Opportunities by FDC structure FDC Structure Fin-Gripper (FESTO. Co) tension Except finger bones, all other parts receive tensile force! FDC Structure in Human Fingers
  • 23. The FDC structure for one-side force direction and curvature can be composed only using soft fabric and belt except one flexible frame. The proposed structure facilitates transfer of tactile sensation because there are soft fabric and belt between the object and human hand. Tension Compre -ssion Tension Tension - Only the right flexible frame receives compressive force. - The others can be replaced by soft materials! Human finger case Tension Moment Assist Flexible frame (ABS+NiTi wire) Belt (Kevlar) Non- stretchable Fabric (Polyester) Only soft materials between the object and the finger!! Wearable FDC Structure for the Moment Assisting Glove 2.2 FDC Structures for Moment Assisting Devices
  • 24. TActily Transparent High-Force Assist Glove (TATH Glove) : 촉감 전달이 가능한 파지력 증강 글러브 2-Finger Test Mechanism Palm side of finger Compression support frame Conduit Kevlar Belt Nitinol Wire Low-stretchable Fabric Conceptual Design of 5-Finger TATH Glove Concept Proof Mechanism Loading Test (7kg) 2.2 TATH Glove using FDC Structure
  • 25. Customized FSR Sensor 2.2 TATH Glove using FDC Structure
  • 26. 전기전자통신공학부 Interactive Robotics & Innovative Mechanism Lab 2.1 Flexible/Soft but Strong? Flexible/Soft and Safe Rigid, Precise and Strong Human hand Flexible/Soft + High payload ?
  • 27. Flexible/Soft and Safe Rigid, Precise and Strong Human hand Precise parallel pinching Parallel pinching Compliant Grasping 손가락은 부드럽게 감아 쥐지만 손끝은 평행한 정밀파지가 가능? Compliant grasping 2.2 Precise parallel pinching + compliant grasping?
  • 28. Parallel pinching Motion Compliant Grasping Motion Fingertip은 수평을 유지하는 동시에, finger body는 flexible한 구조로 물체를 감싸는 동작이 가능함Finger body 유연 파지 Fingertip 평행 핀칭 Triangular structure Parallel structure 2.2 Gripper with Precise Parallel Pinching & Compliant Grasping Soft Parallel Gripper : 정밀 평행 파지가 가능한 유연 그리퍼
  • 29. 생체 이해에 기반한 로봇 핸드 -로봇이 유연성과 정밀성을 모두 가지려면 • 생체 이해에 기반한 로봇 핸드 – RoboRay Hand : 정밀조작과 유연 고하중 파지가 가능한 로봇 핸드 – THAT Glove : 촉감전달이 가능한 파지력 증강 글러브 – Soft Parallel Gripper : 정밀 평행 파지가 가능한 유연 그리퍼 • 생체모사는 형상모사가 아니다. 원리와 기능 레벨에서의 생체 모사가 필요하다 • 유연성과 고하중/정밀성은 trade-off 가 아니다. 양쪽을 다 만족하는 구조 도출 Organic mechanism Fundamental principle Robot Application Organic mechanism Fundamental principle Robot Applications
  • 30. 전기전자통신공학부 Interactive Robotics & Innovative Mechanism Lab 3. 생체 이해에 기반한 로봇 팔 “하이 파이브가 가능한 로봇의 조건”
  • 31. Ave. Inertia : 0.49kgm2 Ave. Mass : 3.9kg Estimated Inertia : over 5kgm2 Mass : 23.9kg Kinetic and potential energy : make a big difference to the safety !! Stiffness/strength : critical to control performance !! 인간의 팔과 로봇 팔의 안전성과 성능 차이 High-five motion 7.13 m/sec Throwing Motion 15.61 m/secPrecise, Strong, but dangerous Inherently safe and still strong, precise and fast VS. Ev = mv2 1 2 Ep = mgh 운동에너지 위치에너지 높은 강도/강성/정밀도  높은 제어 성능 낮은 질량, 높은 역구동성 근본적인 안정성
  • 32. Design Philosophy and Basic Idea Motors with gears in base frame Tendons for joint actuation Reduction gears Motors Motors in base frame Reduction gears Tendons for joint actuation Conventional Mechanism : Stiff but Heavy Tendon Driven Mechanism : light weight but low stiffness Hybrid mechanism : light and Stiff, light-weight gear mechanism is required! 1) Extremely low inertia to minimize stored kinetic energy. 2) Extremely low mass. Conventional industrial robots consume most of the motor torque for supporting their own weight. 3) High stiffness comparable to industrial robots. 4) High strength comparable to industrial robots. 5) Efficiency and backdrivability. In order to apply right amount of energy to the robot, efficient mechanism with minimal friction is required. It enables sensing of external force without expensive force or torque sensors. Stiffness 350Nm/rad Ave. Inertia : 0.49kgm2 Ave. Mass : 3.9kg 근본적인 안전성과 높은 제어성능을 가지기 위한 조건
  • 33. 170 mm 340 mm370 mm 3D-Printed ABS frames Aluminum alloy, machining parts 인간과 동일한 7 자유도 팔 유연파지가 가능한 핸드 3-DOF Wrist 1-DOF Elbow 3-DOF Shoulder 4-DOF Hand 3-DOF Neck Pan, Tilt and Translation 4 Actuators for Wrist and Elbow 3 Actuators with Capstan Reduction for shoulder 무거운 구동모터들 Body 근처에 배치 Joints with Tension Amplification Mechanisms 강성을 증폭하는 경량 감속 메커니즘 LIMS2-AMBIDEX ( Low Inertia Manipulator with High Stiffness and Strength )
  • 34. OVERALL MOTION -Arm span 2.1m -7 DOF per arm -3 DOF per head -Wrist Yaw ROM: 670 deg
  • 35. 35 ARM WEIGHT Total : 2,628 g Wrist : 446 g Forearm : 171 g Elbow : 533 g Upper arm:1,478 g
  • 37. Strength and Stiffness Analysis (LIMS1 Results) z Stiffness: Stiffness/Inertia: >  10,000 Nm/rad 8,333 m/rad 1,410 Nm/rad 7,233 m/rad 350 Nm/rad 2,059 Nm/rad > > • 산업용 로봇과 유사한 동적 제어 성능 – 강성/질량 % Anthropomorphic Low-Inertia High- Stiffness Manipulator for High-Speed Safe Interaction, IEEE Trans. on Robotics, 2017 • 반복 정밀도 0.426mm (산업용 로봇 0.1mm) P2P3 P4 P5 P1 500×500×500mm (a) (b) P2 P3 P4 P5 P1 TABLE V POSITIONING REPEATABILITY TEST RESULTS Pose Mean (mm) Std. Dev. (mm) 3-Sigma (mm) P1 0.138 0.066 0.335 P2 0.148 0.074 0.369 P3 0.103 0.048 0.247 P4 0.187 0.113 0.528 P5 0.189 0.116 0.536 Total 0.153 0.091 0.426 TABLE V POSITIONING REPEATABILITY TEST RESULTS Pose Mean (mm) Std. Dev. (mm) 3-Sigma (mm) P1 0.138 0.066 0.335 P2 0.148 0.074 0.369 P3 0.103 0.048 0.247 P4 0.187 0.113 0.528 P5 0.189 0.116 0.536 Total 0.153 0.091 0.426 • Payload 3kg
  • 38. (a) (b) 0 10 20 30 40 50 0 100 200 300 400 500 600 700 800 900 1000 HIC 36 (m robot ) Effective mass of robots (kg) HIC36 1m/sec 2m/sec 3m/sec 4m/sec 5m/sec 6m/sec 7m/sec 8m/sec 9m/sec 10m/sec 0 2 4 6 8 10 2 3 4 5 6 7 8 9 10 Effective mass (kg) Speed(m/sec) HIC36= 100 HIC36= 200 HIC36= 300 HIC36= 500 HIC36= 1000 1.47kg • HIC (Head Injury Criteria) : a safety criteria derived from the average acceleration of a human head and the application time. mmotor n : 1 m1ink k khead mhead Motor side mass Reducer Link side mass Under 2m/sec, mass is not a dominant factor of safety ( Any mass satisfy HIC100). At 6m/sec, effective mass 10kg can cause almost HIC1000 ! To be safe (HIC100) up to 5m/sec, the effective mass must be under 1.47kg !                  5.2 12 12 , 2 121 1 )(max dtx tt ttHIC t t H tt  HIC 100 : non-life-threatening to the brain, HIC 1000 : life-threatening injuries Influence of Mass and Speed to the Safety 10kg, 6m/sec HIC 958 !! 10kg, 2m/sec HIC 61 Ev = mv2 1 2 0 10 0 100 200 300 400 H 100 200 105 (a) 0 10 20 30 40 50 0 100 200 300 400 500 600 700 800 900 1000 HIC 36 (m robot ) Effective mass of robots (kg) HIC36 1m/sec 2m/sec 3m/sec 4m/sec 5m/sec 6m/sec 7m/sec 8m/sec 9m/sec 10m/sec 0 2 3 4 5 6 7 8 9 10 Speed(m/sec) 1.1kg, 6m/sec HIC 100
  • 41. • Inertia & Stiffness – Shoulder to Hand Inertia : 0.502 kgm2 – Elbow Stiffness : 1,410 Nm/rad • Speed & peak torque – Shoulder roll pitch: 499deg/s, 42.5~82Nm – Shoulder Yaw: 749deg/s, 28.4~54.6Nm – Elbow : 590deg/s, 69.4Nm – Wrist roll pitch: 1179deg/s, 34.7Nm – Wrist yaw: 1634deg/s, 25Nm Specifications Stiffness 350 Nm/rad Inertia : 0.49 kgm2 Mass : 3.9 kg Stiffness 10,000 Nm/rad Estimated Inertia : 5 kgm2 Mass : 23.9 kg Inertia : 0.50 kgm2 Mass : 2.63 kg Stiffness 1,410 Nm/rad AMBIDEX - LIMS2 Human Arm Industrial Robot (iiwa) • Degrees of Freedom – 7 DOF / Arm Shoulder 3, elbow 1, wrist 3 – 3 DOF for Neck Pan, tilt and translation • Weight – Arm moving part : 2.63 kg – Shoulder : 4.17 kg – Hand : 0.633 kg LIMS1 LIMS2-AMBIDEX
  • 42. • Block and Tackle - light weight reduction mechanism 1-DOF Tension Amplification Joint Motor for Wire Motion Rolling Joint Spring Coefficient of wire K Wire Tension Tin Output Tension Tout Actuator Block and Tackle outxinx ,inout nTT  inout x n x  1 in in in out out out Kn x T n x T K 22      Amplified torque Reduced motion Motor for Wire Motion Motor for Wire Motion Motor for Wire Motion Pulley – Symmetric wire motion – High Friction – Limited Range of motion • Tension amplification mechanism for revolute joints Strength : amplified by n times! Stiffness : amplified by n2 times! – Symmetric wire motion – Low Friction – Wide Range of motion ※“An Innovative Trans-umbilical Single-Port Surgical Robot” ICRA2014
  • 43. z 2   w w leftl rightl d • Joint Design Elbow Joint using the 1-DOF Tension Amplification Joint Relationship between wire motion & joint angle Developed Elbow Joint n=6  Stiffness is 36 times higher! Wire pair for wrist actuation Wrist wire length is decoupled with the wrist motion!! 2 sin  nwll rightleft  Symmetric wire motion! 2  n lleft baseleftl _ baseleftl _ 2 leftl 2 leftl al bl cl dl prpr Simplified and actual wire path )2( dactualsimple lrll   Constant The same motion! for even number of winding
  • 44. 2-DOF Tension Amplification Joint Two 1-DOF Joints – Suitable for miniaturization – Frictional wire transmission – Limited workspace due to extended wrist ※“An Innovative Trans-umbilical Single-Port Surgical Robot” ICRA2014 Serial Connection of the 1-DOF tension amplification joint – How to realize spherical rolling contact without slip – Symmetricity of 2 wire pairs should be verified Extension to 2-DOF Mechanism 1-DOF Joint 2-DOF Concept Challenging points z 1-DOF Concept
  • 45. 2-DOF Tension Amplification Joint z Link 1 Link 2 d leftPl _ rightTl _ rightPl _ leftTl _ w a  sinw cosw  sinw cosw b a b a b Bending Plane Symmetric relationship between wires 2 sincos 2 sinsin __ __     nwll nwll rightTleftT rightPleftP   Motion of each wire of the wire pairs is symmetric! Link 1 2p 1p 2t 1t distp Link 2 proxp Ideal Model 0 Link 1 Link 2 1p 2p 2t 1t distp proxp LIMS1 : 구면 rolling 모션 모사 LIMS2 : 3개 link로 단순화, 내구성 향상
  • 46. Link Type Rolling Joint를 이용한 wrist 개선 Wire for wrist actuation : n=4 Stiffness is 16 times higher! Cross roller Planetary gear Universal Joint Wrist 말단부 관절 동력전달 및 감속 구조
  • 47. • 3-finger 4-DOF 핸드 – 파지 물체에 따라 적절히 변형하는 underactuation 구조 – Thumb 자세를 변경하는 추가자유도로 grasping  pinching 변환 가능 LIMS Hand 개발
  • 48. 안전한 로봇 메커니즘 +  괜찮아 들어와.. 정말로 안 아파… 응? 어쩐지 어색해.
  • 49. Summary and Future Work 인간과 협력하는 로봇
  • 50. Summary and Future Work • 생체 이해에 기반한 로봇 팔 “하이 파이브가 가능한 로봇의 조건은?” – LIMS2–AMBIDEX: 안전성과 높은 제어 성능을 가진 로봇 팔 • 생활 속에 로봇이 들어오려면 … Video Ref.: How Smartphones Are Assembled & Manufactured In China. https://www.youtube.com/watch?v=gBL-u53sy_o&t=1s 2013 Artistic Gymnastics World Championships https://www.youtube.com/watch?v=O4pSYZNFej0 The Best Female Rock Climber In the World is 14 Years Old https://m.youtube.com/watch?v=D4zBVD0sL7Y
  • 51. 전기전자통신공학부 Interactive Robotics & Innovative Mechanism Lab Thank you for your attention! 김종인 전형석 정용준 이덕원 장우석 Interactive Robotics & Innovative Mechanism Lab