ME 327: Design and Control of Haptic Systems
Autumn 2015
Lecture 10:
Project/Presentation
Discussion
Allison M. Okamura
Stanford University
presentation
Stanford University
ME 327: Design and Control of Haptic Systems
© Allison M. Okamura, 2015
ME327: Design and Control of Haptic Systems
Autumn 2012
Oral Presentation Evaluation Form
paper presentation
Reviewed By: __________________________________________________
!----------------------------------------------------------------------------------------------------------------------------------Presenter: ______________________________________________________
Paper Title: ____________________________________________________
• select a paper
ME327: Design and Control of Haptic Systems
What are the salient features of the paper? (What was novel or significant about it?)
Slides
Subject
Knowledge
Organization
0
3
Audience cannot
Audience has difficulty
understand the
following the
presentation because
presentation because
there is no order to the the student jumps
sequence in which
around without
information is
connecting different
presented.
topics very well.
What is a good direction for future
work
based
on this
paper?
Student
does
not grasp
Student
is
the information in the
uncomfortable with the
presented paper and
information and is able
cannot answer
to answer only
questions on the
rudimentary questions.
subject.
Student uses
Student occasionally
superfluous graphics or uses graphics that
no graphics; the slides
support the text and
The Presentation
have many errors in
presentation, and/or
spelling, grammar,
the slides have some
and/or
legibility;
or or her
problems
with spelling,
What did this student do particularly
well
in his
presentation?
they are otherwise
grammar, and/or
unprofessional in
legibility.
appearance.
Student mumbles,
Audience members
pronounces terms
have difficulty
incorrectly, and/or
understanding the
speaks too quietly or
presentation. For
too loudly for the
example, student’s
How could this student improve his
or her
presentation?
audience
to hear
well.
voice may be low, and
Alternatively, student
he or she may make eye
may read most of the
contact with the
report and not make
audience only
eye contact.
occasionally.
Student begins late,
Student does not cover
poorly allocates time
all of the material
between topics during
planned and has to
talk, and/or continues
rush, or student fails to
What did you learn from the activity?
speaking long past the
present enough
time limit.
information to fill the
time.
Timing
• prepare your
presentation and
activity
What are the drawbacks of the paper?
Presentation Skills
• sign up for a
presentation slot
see handout for
details
evaluation
The Paper
Autumn 2012
7
Student presents
information in a
reasonable sequence
that the audience can
mostly follow, perhaps
with some effort.
10
Student presents
information in a logical,
interesting order that
the audience can easily
follow.
Student is at ease with
the presented topics
but cannot elaborate on
all issues and does not
handle challenging
questions smoothly.
Student’s graphics
relate to text and
presentation, but there
is room for
improvement.
Slides have few
misspellings, grammar
errors, and illegible
areas.
Student’s voice is clear,
and most words are
pronounced correctly.
Most audience
members can hear the
presentation, but the
student turns to notes
or slides for prompting
several times.
Student demonstrates
full comprehension of
the subject of the paper
and handles questions
adeptly.
Timing is a little off, in
that parts of the talk
feel rushed or slow,
and/or student does
not leave adequate time
for questions and
activity.
Timing is perfect: the
student starts and ends
on time, leaving
sufficient opportunity
for questions and
activity.
Rating
(0-10)
Student’s graphics
explain and reinforce
the text on the slides
and the spoken
narrative. The slides
have no misspellings or
grammatical errors and
look very professional.
Student uses a clear
voice and correct,
precise pronunciation
of technical terms so
that all audience
members can hear the
presentation. Student
maintains eye contact
and does not use notes.
Total
This rating table was adapted from materials by the Information Technology Evaluation Services at the North
Carolina Department of Public Instruction.
Score the presentation on the reverse side.
Stanford University
ME 327: Design and Control of Haptic Systems
© Allison M. Okamura, 2015
project
Stanford University
ME 327: Design and Control of Haptic Systems
© Allison M. Okamura, 2015
project
• decide on team and come up with ideas (due Friday)
3 people; I encourage diversity in experience!
• meet with the instructors to review your ideas
there will be a doodle sign up available this weekend
• propose a specific project (due Nov. 6)
see handout for proposal guidelines
• do the project: make your device/system/experiment
with checkpoints along the way
• present your project at the Haptics Open House
Thursday Dec. 3, 10:30 am to 12 pm in 520-145
your peers, faculty, etc., will be in attendance
Stanford University
ME 327: Design and Control of Haptic Systems
© Allison M. Okamura, 2015
your project should
• be clear in its objectives: know how you define success!
• be informed by a thorough literature search
• be easily used by a haptics novice at the demonstration day
• have high “production values”
• be developed with a user study in mind
Stanford University
ME 327: Design and Control of Haptic Systems
© Allison M. Okamura, 2015
you may wish to...
• have a real impact on the world by solving a problem
(e.g., assistive technology)
• relate the project to your current research
• continue working on your project next quarter as an
ME 391/392 (independent research) project
• make a big splash (publicity, start-up company, etc.)
Stanford University
ME 327: Design and Control of Haptic Systems
© Allison M. Okamura, 2015
example projects from
previous years
(some of these classes had a different focus)
Stanford University
ME 327: Design and Control of Haptic Systems
© Allison M. Okamura, 2015
Display of material softness using
magnetorheological fluid
Stanford University
ME 327: Design and Control of Haptic Systems
2010
© Allison M. Okamura, 2015
Skin stretch haptic device
2010
E. Greenwald, J. Pompe, S. Hsiao, A. Okamura. Quantification and Reproduction of Human Hand Skin
Stretch and its Effects on Proprioception. Understanding the Human Hand for Advancing Robotic
Manipulation (Workshop at Robotics: Science and Systems Conference), Seattle, WA, 2009.
Stanford University
ME 327: Design and Control of Haptic Systems
© Allison M. Okamura, 2015
Tactile display for
simulating lumps in tissue
Stanford University
ME 327: Design and Control of Haptic Systems
2010
© Allison M. Okamura, 2015
Bone screw simulator for
orthopedic surgeons
2010
A. Majewicz, J. Glasser, R. Bauer, S. Belkoff, S. Mears, and A. M.
Okamura. Design of a haptic simulator for osteosynthesis
screw insertion. In Haptics Symposium, pp. 497–500, 2010.
Stanford University
ME 327: Design and Control of Haptic Systems
© Allison M. Okamura, 2015
ented Posture Interface
Xingchi He‡
Department of Mechanical
ngineering, Johns Hopkins
University
Allison M. Okamura
Haptic
feedback based on
kinect tracking
§
2011
Department of Mechanical
Engineering, Stanford
University
be useful in the areas of posture, balance, and rehabilitation, we
have a particular interest in training yoga pose. In classrooms and
in the home, real-time haptic guidance may improve performance
of static pose and reduce yoga-related injury.
1.2 Related Work
Recent work in the area of motion guidance incorporates a variety
of feedback displays and sensing technologies. To highlight a few,
Spelmezan et al. [17] designed a full-body vibrotactile display for
correcting posture in sports that used spatial cues activated remotely
by an instructor; Bloomfield and Badler [8] created a tactile sleeve
for collision avoidance and skill training in virtual environments
that used infrared motion capture; van der Linden et al. [20] developed a wearable jacket with vibration for correcting posture and
violin bowing technique that used inertial motion capture. While all
of these systems use a wearable vibrotactile interface as a “teacher,”
[10], [11], and [16] are most relevant to our work and are further
discussed.
Lieberman et al. [11] designed an upper-body suit with Tactaid
voice coil actuators for vibrotactile feedback for 5-DOF motion
Figure 2: The HAPI Bands system. A user wearing three HAPI
gure 1: The HAPI Bands system helps a user maintain a static
of the arm. Using the Vicon optical tracking system (Vicon MoBands is shown facing a visual display. Four eccentric mass motors
ose.
Five body-worn
bands
vibrotactile
feedback
that
tion Systems,
Los Angeles,
CA), aprovide
user’s motion
was compared
to
are mounted inside each band, and arm bands have accelerometers
uides
the user
to correct
misaligned
torso, shoulder,
and elbow
the desired
trajectory
of a teacher
and vibrotactile
feedback profor limb orientation sensing. The Arduino MEGA microcontroller on
ints.
The
Kinect
sensor,
shown
in
the
inset
image,
is
positioned
portional to joint error was delivered simultaneously to the wrist,
the torso band controls motor activation (blue signals) and reads accelerometer data (black signals). Red connections show power diselbow,
and
shoulder
to
help
the
user
mimic
the
motion.
In
a
motor
front of the user and tracks full-body joint position; band-mounted
M.
F.
Rotella,
K.
Guerin,
X.
He,
and
A.
M.
Okamura.
HAPI
Bands:
A an
Haptic
Augmented
Posture
Interface.
tribution
from
external
supply, and purple
signals
represent In
USB
task, tactile feedback
improved
performance
by 27%. ASergi
et al. of the
ccelerometers
measure
upper-limb
orientation.
skeleton
Proceedings
IEEE
Haptics
Symposium,
pp.
163-170,
2015.
communication
between
Kinect,
computer,
and
microcontroller.
[16]pose
focused
regulatingon
the an
orientation
of the
forearmwith
in Carteser’s
is on
displayed
external
screen,
coordinate
sian space, using the Liberty magnetic tracking system (Polhemus,
ames
indicating
orientation of upper-body
joints.
Stanford
University
ME 327:
Design
of Haptic Systems
© Allison M. Okamura, 2015
Colchester,
VT). Tactile feedback was delivered
to the
wristand
via Control
an
instrumented bracelet with four motors. Vibration proportional to
arm rotation. Each band is constructed from two-inch wide elastic
Haptic feedback of prosthetic hand
configuration
2011
!
C0
!
C1
C2
C3
C4
C5
!
A. Cheng, K. A. Nichols, H. M. Weeks, N. Gurari, and A. M. Okamura. Conveying the Configuration of a Virtual
Human Hand Using Vibrotactile Feedback. In Proceedings IEEE Haptics Symposium, pp. 155-162, 2015.
Stanford University
ME 327: Design and Control of Haptic Systems
© Allison M. Okamura, 2015
2011
Haptic feedback for virtual
keyboard/buttons
!
!
!
Stanford University
ME 327: Design and Control of Haptic Systems
© Allison M. Okamura, 2015
2011
Haptic paddle juggling
t*TJUDSJUJDBMGPSQFSGPSNBODF
t3FEVDFETUBUFTQBDFEJNFOTJPO
5SBJOJOHBOEEBUBDPMMFDUJPOTFTTJPOT
)BQUJD
1BEEMF
Percentage of Outlier Apex Heights
Haptic Feedback
No Haptic Feedback
50
CV [%]
PO [%]
40
30
20
10
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Subject #
M. M. Ankarali, H. T. Sen, A. De, A. M. Okamura, and N. J. Cowan. Haptic Feedback Enhances Rhythmic Motor
Control By Reducing Variability, Not Convergence Rate. Journal of Neurophysiology, 111:1286-1299, 2014.
Stanford University
ME 327: Design and Control of Haptic Systems
© Allison M. Okamura, 2015
AngryGrads: Haptic feedback for
learning a dynamic task
Stanford University
ME 327: Design and Control of Haptic Systems
2011
© Allison M. Okamura, 2015
2011
Foot haptics
!
!
!
!
R. P. Jayakumar, S. K. Mishra, J. F. Dannenhoffer, and A. M. Okamura. Haptic Footstep Display. 2015 IEEE Haptics
Symposium. In Proceedings IEEE Haptics Symposium, pp. 425-430, 2015.
Stanford University
ME 327: Design and Control of Haptic Systems
© Allison M. Okamura, 2015
A 2-DOF Haptic Device for Displaying
Forces on an Airfoil
2014
Appendix
Screenshot of the user interface:
Arduino Code:
Stanford University
ME 327: Design and Control of Haptic Systems
© Allison M. Okamura, 2015
y
s.
h
A-B design to eliminate order effects, the subject performed these
movements without skin-stretch feedback in blocks 1 and 3 and
with skin-stretch feedback — at the index fingertip, according to
4 R ESULTS
the control paradigm believed to be most intuitive — in blocks Figure
2
4 displays the results from the anonymous poll on skinand 4. Each block, which lasted 6-8 minutes, was followed by stretch
a
“intuitiveness”. Those who responded to the poll were
3-minute break. The first session of blocks was preceded by a 15mainly Stanford engineering students and professors. Greater than
80% of users found the anti-aligned control paradigms to be more
minute training session with the Emotiv Control Panel (to build the
intuitive than their aligned counterparts. Of these users, over 65%
EPOC’s neural-activity mapping) and a 10-minute period of feelpreferred position-based over velocity-based control; this repreing the different skin-stretch control paradigms. This protocol was
sented nearly 60% of all users polled.
approved by the Stanford University Institutional Review Board.
Figure 3 shows our subject performing the study.
Skin-Stretch Haptic Feedback for
Improved Control of Brain-Computer Interfaces
n-
12
ne
sunt
N umb e r of Vot e s
10
e
e
sd
ur
i+)
le
s
d
ll
2014
8
6
4
2
0
pos–
vel–
vel+
pos+
C ont r ol Par adigm
3: Subject
Performing
Study
S. M. Sketch, Figure
D. R. Deo,
J. P. Menon
andthe
A. User
M. Okamura.
Design and
experimental
ofControl
a skin-stretch
Figure
4: Intuitivenessevaluation
of Skin-Stretch
Paradigms
haptic device for improved control of brain-computer interfaces.
the BCI-basedpages
study, 272-277,
our expert subject
In IEEE International Conference on Robotics andFor
Automation,
2015. was a 22-year-old,
healthy, right-handed male with no sensory deficits. Tables 1 and
an hour of training with the Emotiv EPOC, our subject was able
2 display
his performance for each ©
session
block 2015
of the
Stanford
University
MEdirection
327: Design
Control
of Haptic
Systems
Allisonand
M.each
Okamura,
to move
the cursor in a pre-commanded
with and
80-90%
accuracy.
BCI-based study, both in terms of the number of target acquisitions
6 After
project discussion
and matchmaking
Stanford University
ME 327: Design and Control of Haptic Systems
© Allison M. Okamura, 2015