Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Workshop_HUMANOIDS2014

Final version
by

Markus Grebenstein

on 13 June 2017

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Workshop_HUMANOIDS2014

Motivation
Why the State of the Art Matters:
The Importance of Functional Abstraction in Robot Hand Design

Markus Grebenstein, Madrid 18.11.2014
Design
Actuation
Conclusion
Functional understanding of characteristics and strategies

Merge assets of biological and technical systems
Imitate human anatomy

Use only assets of biological system

Pros and cons of biological system + losses due to technical realization
Methodology
Requirements:
Hardly inertia => Static forces

Design space => Packaging

Complex system => Simplicity

Drives in forearm => Tendon driven

Exposed => Robustness
Results
Dynamics
Understand
Copy
2 DOF

4 bar mechanism (3D antiparallelogram)

38 tendons

6 kN tendonload (worstcase 10kN)
Mechanics
Fingers
PIP/DIP Joint / Thump MP/IP
Metacarpal Joint/ Thumb TMC
Wrist
Housings
Antagonistic
Drives
[Grebenstein ICABB 2010]
Kinematics
Redesign/ Measures
Prototypes
Measurement of human kinematics
[Guirintano 1995], [Stillfried 2010]
Kinematics optimization
[Santos 2004]
There is no optimal kinematics but
almost 7 billion well working ones
Tests
Use proven knowledge: Hand surgeon tests
Use human intuition: Grasping tests
Play
around
[medscape.com]
State of The Art
Empirical kinematics
Robot Hands
Human
Hand Analysis

Palm
Context
Anatomy
Finger IP Joints
Inclination:
Angle between sagital plane and joint axis within frontal plane
!not really!
Functionalities
Design Guidelines
[Leinsje, et. al; 2008]
[Mc Grouther et al. 2000]
[Tubjana 1993]
[Mc Grouther et al. 2000]
[www.wikipedia.de]
Friction Estimation
Guidelines
Functionalities
"Anatomy"
Twendy One
MIT Hand
Armar
Robonaut
MIT Hand
Twendy One
Robonaut
Armar
Keys

Fingers need to wrap object =>DIP is a necessary joint
MC and TMC/HMC Joints
Ligaments
Twist:
Angle between subsequent axes around longitudinal axis of the bone
MC condyloid joint functionality is representable by a cardan joint

First axis of MC joint has to point out of the palm

Inclination of finger joints is crucial to grasp spherical objects

Twist of the thumb IP adjusts thumb pad orientation

Do not introduce couplings if you do not know which functionalities are affected/needed
Forearm
[Friedl et. al 2011]
Crossed routing vs. parallel routing
Hyperboloid vs. cylinder
Control
Transmission
Actuation Principle
Individually adapted spring characteristics

Intrinsic tensioner
Parameter space is too large=> optimization of single fingers

Cost functions, mainly derived from robot arms, are not proven for robotic hands
Manipulability
Dexterity
Grasp quality
...

Completeness of any designed task set is uncertain i.e. regarding manipulation
Contradictory Tests
Dyneema Tendons
durable
spliceable
colored
notable creap
Thumb Kinematics
Placement and orientation of thumb base (TMC)
[Grebenstein HUMANOIDS 2010]
Tendon Routing
Tendon Routing/
Assembly Concept
Central Palm
Mechanics
Palm Mechanics
Details
Functionalities
Force transfer
Stress distribution

Concepts
Glove
Composite shells
Robustness
Grasp Robustness
Full System
Experiments
Testbed Experiments
Impact energy stored in springs:
Motor acceleration phase >2.5 times

Motor speed 1/6 joint speed
Singularity
No twist
Twist
Joint velocity and joint torque
MC
PIP
DIP
MC
PIP
DIP
Ext. 1+2
Ext. 1
Ext. 2
Flex.
Ext.
Flex.
Ext.
Flex.
Ext.
Flex.
Ext.
PIP
MC
Input
E>6 J

v>4 m/s

m=0.75 kg
Results:
Maximum joint speed: >1000°/s

Object impact force: appr. 900 N

Maximum tendon force: <70 N
Objective
Introduce the idea of functional abstraction

Apply it to human and robotic hands

Derive some example design guidelines
Finger flipping
appr. 4000°/s


Human (little dynamic preloading):
MC approx. 1000°/s
PIP approx. 690°/s [Darling 1990]
[www.santafeeats.com]
Saddle Joint
Condyloid Joint
Hinge Joint
MC Joint Functionality
IP Joint Functionality
[Grebenstein 2010]
Orientation and order of TMC axes

Placement of first TMC axis singularity
Flex.
Ext.
You better know what others did....

Functionally
Thanks for your attention
[www.nasa.gov]
[Jacobsen et al. 1984]
[www.nasa.gov]
[Sugano Labs]
[Kargov et al. 2005]
19 DoF
[Grebenstein 2011]
[Butterfass et. al.1998]
[Jacobsen et al. 1984]
[Mouri et al. 2002]
[Deshpande 2013]
[Kapandji 2001]
[Diftler et. al. 2011]
Gifu Hand 3
Grasping Performance
Full transcript