Conference PaperPDF Available

Educational Results of the Personal Exploration Rover Museum Exhibit


Abstract and Figures

The Personal Rover Project produces technology, curriculum and evaluation techniques for robotic educational use in formal and informal (after-school, out-of-school) learning environments. Our specific aim for this phase of the project is to create and evaluate human-robot interactions that educate members of the general public in an informal learning environment, specifically museums. Our educational goals are to further an appreciation and understanding of NASA's Mars Exploration Rovers (MERs), to illustrate the role of robotic rovers in scientific exploration, and to provide hands-on learning experiences that demonstrate robot autonomy. We have designed a new robot, the Personal Exploration Rover (PER) and the related interactive components of a museum exhibit to achieve these goals. Here we describe the exhibits developed and the formal evaluation results of the exhibits' educational impact and efficacy. These results suggest techniques by which learning can be measured and used as an indicator of successful human-robot interaction.
Content may be subject to copyright.
Educational Results of the Personal Exploration Rover
Museum Exhibit*
Illah Nourbakhsh, Emily Hamner, and Brian
The Robotics Institute
Carnegie Mellon University
Pittsburgh, PA, USA
Debra Bernstein and Kevin Crowley
Learning Research and Development Center
University of Pittsburgh
Pittsburgh, PA, USA
Abstract*- The Personal Rover Project produces
technology, curriculum and evaluation techniques for robotic
educational use in formal and informal (after-school, out-of-
school) learning environments. Our specific aim for this
phase of the project is to create and evaluate human-robot
interactions that educate members of the general public in an
informal learning environment, specifically museums. Our
educational goals are to further an appreciation and
understanding of NASA's Mars Exploration Rovers (MERs),
to illustrate the role of robotic rovers in scientific
exploration, and to provide hands-on learning experiences
that demonstrate robot autonomy. We have designed a new
robot, the Personal Exploration Rover (PER) and the related
interactive components of a museum exhibit to achieve these
goals. Here we describe the exhibits developed and the
formal evaluation results of the exhibits' educational impact
and efficacy. These results suggest techniques by which
learning can be measured and used as an indicator of
successful human-robot interaction.
Index Terms – Educational Robotics, Human-Robot
Interaction, Social Robots, Robot Autonomy
The need for significant improvement in technology
literacy and education cannot be overstated, especially as
technology becomes increasingly present in day-to-day
human activities. Yet, in recent years, computer science
and engineering departments have begun to suffer from
declining student enrollment. We and other researchers
have begun to explore the role that robotics can play in
engaging and retaining students in technology-related
curriculum and fields [2], [7], [12], [15], [19], [21]. This
agenda has been furthered by our and others’ results
which show that educational robotics can trigger
significant learning across broad educational themes that
extend well beyond STEM (science, technology,
engineering and mathematics) and into the associated
lifelong learning skills of problem-solving, collaboration
and communication [6], [9], [11], [16], [20], [22].
The Personal Rover Project, a multi-year educational
robotics study, has focused specifically on the application
of interactive, physically embodied robotic technology for
education in formal and informal learning environments
[8]. The educational goals of the project as a whole are:
Inspire students to explore boundaries of their
knowledge and creativity through the use of science and
* This work was funded by NASA/Ames and Intel Corporation.
technology and to pursue careers in math, science and
Stimulate public awareness and interest in the NASA
mission and reveal the challenges associated with using
robotic devices for science and exploration.
Teach children the critical skills of teamwork,
collaboration, problem-solving and inquiry-based science.
Prior stages of the Personal Rover Project have
identified design principles for the creation of richly
expressive low-cost robotic platforms [10], and have
deployed educational robotics curriculum for structured,
formal learning environments [16]. Results derived from a
formal analysis of a robotic autonomy summer course for
high-school students included significant improvement in
the sought-after area of retention of girls in technology-
intensive coursework.
This paper describes the most recent endeavor of the
Personal Rover Project, the creation and evaluation of a
robot-based exhibit in informal learning environments that
features unmediated, short-term human-robot interactions.
In this phase of the project we focused on learning in
informal settings where total time on task between visitors
and robotic technologies is measured in minutes rather
than days. Our goal was to assess whether robotic devices
could offer significant educational advantages even in
transient human-robot interactions, such as those
experienced in a museum visit by tens of millions of
visitors per year.
Motivated by the expected broad exposure and public
interest in NASA’s Mars Exploration Rover (MER)
missions targeted to land in January 2004, we elected to
launch a technology-based educational experience related
to the MER missions in a number of science museums and
technology centers across the country. Visitors would
interact with the Personal Exploration Rover (PER), a
robotic science rover (Fig. 1) via a kiosk-based ‘mission
control’ interface to identify and search Martian rocks for
signs of organic life.
The PER robot was designed to meet its specific
educational objectives within the context of the NASA
MER missions. Two key objectives were:
Show that rovers are tools for doing science by enabling
visitors to act as mission scientists, and use the PER robot
to conduct a science operation.
Enable visitors to appreciate the role of autonomy on
board rovers.
Proceedings of the 2005 IEEE
International Conference on Robotics and Automation
Barcelona, Spain, April 2005
0-7803-8914-X/05/$20.00 ©2005 IEEE.
Fig. 1. A P ER tests a rock for life at t he National Scien ce Center .
Museums are prime venues to evaluate these objectives
because they offer human-robot interaction opportunities
over a sufficiently large body of diverse visitors such that
statistically meaningful conclusions regarding interaction
and education can be drawn.
In just the first four months, PER robots effected more
than 20,000 autonomous science target approaches and
completed greater than 30 miles of rover travel with
minimal operating failures. A detailed description of the
PER robot can be found at [17] and [18].
The PER exhibit installations present museum visitors
with the challenge of using the PER to search for signs of
life on rocks placed in a Martian terrain sandbox called
the “Mars yard”. A PER mission starts with the rover
compiling a 360° panoramic image of the surrounding
yard. At the ‘mission control’ kiosk located outside the
Mars yard, users are presented with the panorama in the
“Mission Builder” screen (Fig. 2) that guides them
through the planning of a rover mission.
Fig. 2. The “Mission Builder” screen display.
First, users must interpret the panoramic image data and
select which target rock to send the rover towards to
search for signs of life. This first selection provides the
rover orientation information for the mission. Second,
users must locate the position of the rover and the selected
target rock on an orthographic, overhead satellite map
image. Together these provide sufficient rover angle and
distance information to complete the mission. To help the
user orient between the physical Mars yard and the
onscreen panoramic display of the yard, a Martian sun is
painted on the far wall of the yard, visible both from the
kiosk and on the screen (Fig. 3). In addition, the rock
positions, rock shapes, and the shape of the yard help
users interpret the satellite map [17].
Fig. 3. The ability to see the yard and kiosk screen simultaneously aids
users in orienting themselves within the exhibit.
As the rover executes the mission, a rover’s-eye view
camera allows visitors to experience the mission from the
rover’s perspective via real-time video. The “Rover
Mission” sub-window at the bottom right of the “Mission
Builder” screen remains visible during execution of the
mission, providing data regarding rover operations,
distance traveled and angles turned. Along the way the
rover continually scans for obstacles in its path using an
IR rangefinder mounted in its pan-tilt head. After the
rover has turned and driven the distances specified, it
demonstrates further autonomous capabilities by scanning
for the target with the same IR rangefinder, determining if
a target can be located, refining its position and alignment
with the target rock, and performing an ultraviolet test for
signs of life (Fig. 1). Some of the target rock faces are
painted with an invisible fluorescing paint that glows
under ultraviolet light. An image of the illuminated target
rock is returned to the kiosk for scientific analysis by the
user. The entire interaction is designed to be completed
easily within three minutes to satisfy throughput
requirements for high-traffic museums. An end message
and countdown screen signal mission completion and
allow the next user to step in.
The PER exhibit to date has been deployed at five main
locations across the country: the Smithsonian Air and
Space Museum (NASM), the Smithsonian Udvar-Hazy
Center, the San Francisco Exploratorium, the National
Science Center, and the NASA Ames Mars Center. The
exhibit installations opened between December 29, 2003
and January 24, 2004, to coincide with the landing of the
MERs on Mars, and each ran for two months or more. As
of January 2005 the Udvar-Hazy and NASA Ames
exhibits continue to operate and the exhibit is scheduled to
open at the Japan World Expo in March 2005.
A. Exhibit format
The presentation format for the exhibit is left up to the
individual museum. As a result we have observed three
different styles of exhibit interaction—mediated, semi-
mediated and unmediated. At NASM a docent is stationed
next to the ‘mission control’ kiosk, in order to provide
information about the PER (and MER mission) and to
guide the visitors through a mission. At the Hazy Center,
the exhibit is used for structured teaching activities with
school groups. We consider these types of interaction
mediated as a human assists the user in the human-robot
interaction. At the Exploratorium, National Science
Center, and NASA Ames families interact with the PER
robot exhibit on their own in an unmediated fashion,
although staff members are generally available if needed
to answer visitor questions.
B. Mars yard designs
Each museum designed and produced its own Mars yard
or yards for the exhibit. The Mars yard terrain and
topography are specifically designed with the PER’s
physical capabilities and the desired exhibit interaction in
mind. The rocks, rubble and hills in the yard are all
traversable by the rover except for four or five large rocks
that serve as scientific targets. The yards are encircled by
low walls portraying real Martian landscape and horizon
imagery from NASA’s Pathfinder mission. Each yard also
features a sun on one wall designed to help the visitors
orient themselves when using the exhibit. The hip walls
are of sufficient height to be viewed as obstacles by the
rover but low enough to allow visitors a view of the yard
and rover.
The National Science Center and Exploratorium each
have two yards, while the other locations each have a
single yard. The dimensions and shapes of the Mars yards
vary based on the space and material constraints of each
location. The largest yard is at the NASA Ames Mars
Center and measures 16 feet on each side. The smallest
yard is approximately 8 feet by 9 feet. At the National
Science center, the yards are polygons designed to
maximize available space. The yards are constructed from
spray painted Styrofoam; layered paint, glue, sand, wood
and plaster; small lava rocks and sand; and layered
Styrofoam, polymesh and dryvit compound.
Fig. 4. This picture of the NASM yard was taken during installation of
the exhibit, before the horizon images were added. The yard is built on
casters and designed to split into quarters so that it can be easily moved.
Local high school students built both the NASM and
Hazy Center yards. Using data from the Pathfinder
mission, Earth Science classes designed the yard
topography from cut Styrofoam to be an exact scale model
of real Martian terrain. Art classes covered the foam with
dryvit, painted the yard, and built realistic looking
Styrofoam rocks. The end result is a realistic Martian
terrain for the PERs to explore (Fig. 4).
Quantitative statistics regarding exhibit use were
collected automatically at installations by the exhibit
software itself and by sampled passive observation. Both
quantitative results and informal observations guided the
more formal educational exhibit evaluation that followed.
These statistics identify the demographics of the exhibit
users and the manner in which the exhibit was used.
Significantly, the statistics show that time on task is
extremely close to the design target of 3 minutes and more
importantly virtually all exhibit users were able to
successfully complete the entire mission. Together these
statistics indicate that the distribution of time on task is
not, as is often the case in museum exhibits, exponential
but rather unimodal and narrow. Users who are engaged
by the PER exhibit remain engaged through mission
completion, then helpfully release control to the next
museum visitor in queue. Details of both user
demographics and mission use statistics follow.
A. Audience
Exhibit use observations were conducted at the
Exploratorium and NASM. At both locations, the exhibit
was in nearly constant use. Over roughly 4.5 hours of
observation, 184 people interacted with the exhibit. This
included 71 adult users (36 females and 35 males), and
113 child users (28 females and 85 males). The majority
of exhibit users were in groups, and the average group size
was 3.06 (σ 1.22), with a total of 64 groups using the
exhibit during this period. Group members often took
turns conducting rover missions. Although more boys than
girls were present at the exhibit, 61% of boys and 71% of
girls attending the exhibit operated the rover.
B. Mission statistics
Based on logs automatically generated by the
Exploratorium and NASA Ames kiosks between Dec 29th,
2003 and April 14th, 2004 we are able to report additional
information about exhibit use1. The exhibits were in use
75.4% of the time while they were open (331 hours idle
and 1017 hours in use). Out of 26,200 missions only 525
(2.0%) timed out before the end of the Mission Builder
screen, meaning that 98% of users were able to
successfully design a mission and send it to the rover. This
represents a surprising retention statistic, in that users tend
to engage and stay with the PER through an entire
mission, virtually never leaving early. When a mission is
1 All of the kiosks generate logs, but these results are based upon
NASA/Ames and Exploratorium analyses only.
unsuccessful, users are given the option to try again or
quit. Only 499 (1.9%) missions timed out at this stage,
showing that users were highly engaged even when their
mission failed to find the target rock. The average mission
length was approximately 2 minutes 20 seconds (139.7
seconds σ 60.1 seconds). This is the length of time for a
single set of instructions to be selected by the user, sent to
the rover, and executed. On average each user engaged the
PER in 1.6 missions (σ 0.94), thus the overall individual
time on task is approximately 4 minutes, exceeding the 1.4
minute engagement time typically seen at interactive
science exhibits [5].
About half of the missions (52.7%) ended with the rover
successfully locating a rock (Fig. 5). The next most
common outcome (23.1%) was that the rover was blocked
by an obstacle, generally a rock, while it still had more
than 150 cm left in the path to its target. The rover went
“out of range”, i.e. detected a hip wall blocking its path,
only 18.1% of the time. In 3.4% of the missions, the
mission ended due to a robot error such as failed
communication. Because the IR sensor’s range is limited
to 150 cm, the rover was unable to locate any rock or hip
wall 2.7% of the time.
In summary it is clear both from time on task values,
time-out rarity and mission success rates that visitors are
able to effectively use the PER exhibit, even in the
unmediated cases of the Exploratorium and NASA/Ames
installations. It is further clear that for children, there is no
obvious statistical gender gap in terms of engagement
with the PER exhibit. Both of the above conclusions are
hopeful in that the PER exhibit attracts and engages the
target population. The ability of the PER exhibit to engage
and retain the interest of girls as well as boys is
noteworthy; this echoes earlier educational results from
the Robotic Autonomy summer course [16]. The next
question, addressed in the following section, is whether
this exhibit uses technology in an educational manner.
Mission Results at NASA Ames and the Exploratorium
0% 10% 20% 30% 40% 50% 60%
No Rock
Fixed Error
Fatal Error
Out of Range
Mission Results
Percentage of Missions
Fig. 5. Mission results from NASA Ames and the Exploratorium
between December 29th , 2003 and April 14th, 2004.
The Learning Research & Development Center
conducted formal educational evaluation of the PER
exhibit at NASM and the Exploratorium. These two
museums were chosen as research sites in order to provide
a full picture of how the exhibit functioned with different
levels of museum mediation. The goal of the evaluation
was to see if people were engaging with the intended
content of the exhibit.
Traditional school-based assessments of learning are
often inappropriate for use in informal learning settings
[1]. As groups of visitors use and talk about exhibits, they
are constructing a shared understanding of the content.
Following recent theoretical and empirical work in
museum learning [4], [14], our analyses focus on this
naturally occurring talk as the best indicator of whether
the exhibit successfully meets its educational goals.
In this article, we focus upon one of the most common
exhibit user groups: children visiting the museum with
families. We first analyze videotapes of families using the
exhibit in order to describe the extent to which their
conversations reflect the intended educational themes.
Second, we analyze post-exhibit interviews with children
in order to describe the extent to which they understood
those same themes after having used the exhibit.
The post-exhibit child interview consisted of a set of
open-ended questions about the Mars mission, the MERs,
and the PERs. Here we present selected analyses focusing
on the questions of how the exhibit supported the two
educational objectives of allowing visitors to explore 1)
the role of robots in mission science and 2) the nature of
robot autonomy. For additional analyses see [17].
A. The Role of Robots in Mission Science
In this section, we evaluate the extent to which the
exhibit supported visitor learning of the role of robots in
scientific exploration. Fig. 6 presents the percentage of
conversational groups2 discussing different topics. These
data suggest that the PER exhibit supported conversations
about the Mars mission and general robotics at both sites.
However, conversation groups at NASM, which included
a docent, were significantly more likely to talk about the
Mars mission and to make explicit comparisons between
the MER and the PER.
Further analysis of the conversation data revealed that
parents generally initiated the same amount of thematic
talk at both the Exploratorium and NASM exhibits3, and
that the docents seem to be responsible for the increase in
the amount of thematic talk at NASM. However, even if
docents are able to provide additional factual information
to museum visitors, is it not necessarily the case that the
PER exhibit is more successful when mediated by a
docent. Anecdotal evidence suggests that parents are more
sensitive to their children’s understanding of the rover,
and are more likely to tailor their comments at the exhibit
2 As a unit of analysis, the conversational group includes anyone present
at the exhibit with the child. At the Exploratorium, the conversational
group generally included the child, parent(s), siblings and any other
exhibit users with whom the child interacted. At NASM, the
conversational group included the child, parent(s), siblings, other exhibit
users, and a docent.
3 With the exception of talk about collaboration with robots (i.e., people
and robots working together to solve problems), which was initiated
more often by parents at the Exploratorium.
to correct children’s misperceptions. Research in the field
of museum learning suggests that parents can serve an
important bridging function between children and
museum exhibits, making the parent’s role an important
one [3].
When interviewed after exhibit use, almost all children
at both the Exploratorium and NASM possessed basic
knowledge of the Mars rover mission (93% and 100%
respectively). Additionally, 21% of children at the
Exploratorium and 38% of children from NASM made
spontaneous comparisons between the MER and the PER.
With regard to person-robot communication, 72% of
children at the Exploratorium and 69% of children at
NASM were able to describe the devices people can use to
communicate with robots (e.g., computers and, in the case
of rovers in space, satellites). There were no statistically
significant differences between children from NASM and
the Exploratorium for any of the categories reported here.
Themes Exploratorium NASM
About the Mars Mission* 55% 93%
Comparisons between MER and PER* 24% 79%
Communicating with Robots 45% 72%
Collaborating with Robots 86% 93%
*indicates a statistically significant difference between the
Exploratorium and NASM groups, p<.01
Fig. 6. Percentages of conversation groups at each museum discussing
themes related to the role of robots in mission science.
B. The Nature of Robot Autonomy
This exhibit was designed to provide museum visitors
with the knowledge and information necessary to
appreciate the importance of rover autonomy. Although
all museum visitors come to the exhibit with prior ideas of
what robots are and what they can do, most have probably
not interacted with a robot that possessed true autonomous
properties [13]. Thus, the exhibit experience provides a
unique opportunity for visitors to re-evaluate their
concepts of what a robot is and what a robot is capable of
Fig. 7 shows the percentage of conversational groups
discussing different topics at the Exploratorium and
NASM. Conversational groups at both museums
addressed each topic, although all topics were addressed
significantly more frequently at NASM. Analysis of the
source of exhibit conversation revealed that parents at
both the Exploratorium and NASM discussed these topics
with similar frequency. As was the case in the previous set
of exhibit conversation analyses, the docents seem to be
responsible for the increase in frequency of thematic talk
at NASM.
Themes Exploratorium NASM
Rover Design* 34% 93%
Rover Activities* 45% 100%
Rover Autonomy* 52% 93%
*indicates a statistically significant difference between the
Exploratorium and NASM groups, p<.01
Fig. 7. Percentage of conversation groups at each museum discussing
themes related to rover autonomy.
In order to assess children’s ideas about rover
capabilities, children’s interview transcripts were coded
using two categories: rover design and rover activities.
Children from both the Exploratorium and NASM were
able to speak knowledgably about the technology on the
rovers and the type of actions they were capable of
performing. Fifty-two percent of children from the
Exploratorium and 77% of children from NASM talked
about rover design (e.g., the technology typically found in
rovers, such as motors, cameras, range finders). Similarly,
55% of children at the Exploratorium and 85% of children
at NASM were able to describe the types of activities a
rover could perform (e.g., taking pictures, driving,
exploring); this difference was marginally significant, X2
(1, N = 42) = 3.39, p=.06.
Assessing children’s ideas about rover autonomy
proved to be more challenging, as some children were
inconsistent or unsure of whether a robot would be
capable of autonomous behavior. To address this issue, we
devised a separate system to measure both the adequacy
and the strength (consistency) of children’s ideas about
robotic autonomy. For each statement indicating an
understanding of the autonomous operations of the rover,
children were given one positive point. For each statement
indicating the opposite belief, namely that the rovers were
incapable of independent action and operated via remote
control, children were given one negative point. This
system was applied to children’s answers to open-ended
questions about how the rovers operate.
Neither PER nor MER autonomy scores correlated
significantly with age, although, as one might expect, PER
and MER scores were significantly correlated with each
other (r=0.48, n=42, p=.001). Across institutions, there
were no significant differences in PER or MER autonomy
In total, over 40% of children at the Exploratorium and
over 45% of children at NASM left the exhibit with some
understanding of the autonomous capabilities of the PER.
Similarly, 31% of children at the Exploratorium and over
50% of children at NASM came away understanding the
autonomous capabilities of the MER4. The somewhat
higher autonomy scores at NASM may be a function of
the explicit conversation from docents regarding robot
autonomy. Cognizant of the goals of the exhibit, docents
were more likely to be explicit in their descriptions of the
rover’s autonomous behavior than were parents. It may be
the case that for a concept as difficult as robotic
autonomy, children benefit from explicit descriptions and
definitions of autonomous behavior.
C. Analysis Conclusions
This assessment suggests that the exhibit was successful
in meeting its core goals of involving visitors in
explorations of the role of robots in mission science and of
robots as autonomous entities. Analysis of family
conversation suggests that visitors were expanding on
relevant themes as they used the exhibit. They talked
4 These percentages represent the number of children with positive
autonomy scores for the MER and PER.
about the ongoing Mars mission, compared the MER and
PER, discussed communicating and collaborating with
robots, and talked about robot design, technology, and
autonomy. Interviews with children following the exhibit
suggested that almost all children were aware of the Mars
mission and that many of them also were able to connect
the exhibit experience in specific ways to the mission.
Children did not end their experience with a uniformly
robust view of autonomy. Although some recognized
autonomous characteristics of the rovers, most children
held inconsistent theories. More than half still held views
that the rovers are primarily operated through direct
remote-control. We do not necessarily believe that a single
exhibit experience would be a sufficient base for children
to develop fully correct theories of autonomy. The exhibit
experience is probably best seen as a chance for families
to work out some of these issues in the context of an
authentic autonomous rover. Still, future versions of such
exhibits should be designed to more explicitly challenge
children’s incorrect or inconsistent theories.
The Personal Exploration Rover has served as a
rewarding demonstration of educational robotics applied
to the informal learning space. Given concrete goals in
relation to the NASA Mars Exploration Rover mission,
this team designed a new educational rover and graphical
interaction system, installed the resulting exhibit at
multiple high-traffic museums across the country, and
performed quantitative and qualitative evaluation of the
exhibit’s efficacy. In summary this project demonstrates
that robotic technology has compelling value in the
museum setting, and that concrete educational results can
be achieved and measured in such a setting. Exhibit
statistics suggest that, among children, girls and boys are
both engaged by this robotic exhibit, to such a degree that
virtually all users succeed in the completion of an entire
scientific rover mission. Educational evaluation suggests
that the exhibit effectively serves as a platform for family
discussions about the MER mission and robotics, and that
children come away from the exhibit with measurable
knowledge in these areas. These results also indicate that
learning can be evaluated and used as a critical measure of
successful human-robot interactions.
As robotic technology advances, future teams will be
capable of creating ever more compelling exhibits and
curricula for both formal and informal learning venues.
We hope that this project can serve as a motivation for
future teams to not only research, dream and invent, but
also to harden, fabricate and install so that thousands can
benefit from these educational technology ventures.
We would like to thank Ellen Ayoob, Doug Baldwin, Jim
Butler, Corinne Cannon, Daniel Clancy, Jeff Cross, Maylene
Duenas, Joe Edwards, Edward Epp, Jim Frye, Rachel Gockley,
Jean Harpley, Robyn Higdon, Thomas Hsiu, Mark Lotter, Marti
Louw, Andrew McClellan, Nicole Minor, Victor Morales,
Jennifer O'Brien, Anuja Parikh, Eric Porter, Mike Reeves, Skip
Shelly, Kristen Stubbs, Tom Roach, Priscilla Strain, Nick
Veronico, Noel Wanner, Ollie Washington, Steven Williams,
Peter Zhang, and Cheryl Zimmerman.
[1] Allen, S. Looking for learning in visitor talk: A methodological
exploration. In G. Leinhardt, K. Crowley & K. Knutson (Eds.)
Learning Conversations in Museums (pp. 259-303). Mahwah, NJ:
Lawrence Erlbaum Associates (2002).
[2] Beer, R., Chiel, H., & Drushel, R. Using autonomous robots to
teach science and engineering. Communications of the ACM, June
[3] Crowley, K. & Callanan, M. Describing and supporting
collaborative scientific thinking in parent-child interactions.
Journal of Museum Education, 23, 12-17 (1998).
[4] Crowley, K., Callanan, M., Jipson, J., Galco, J., Topping, K. &
Shrager, J. Shared scientific thinking in everyday parent-child
activity. Science Education, 85(6), 712-732 (2001).
[5] Crowley, K., Callanan, M., Tenenbaum, H. & Allen, E. Parents
explain more often to boys than to girls during shared scientific
thinking. Psychological Science, 12(3), 258-261 (2001).
[6] Druin, A. & Hendler, J. Robots for kids: exploring new
technologies for learning, The Morgan Kaufmann Series in
Interactive Technologies, Morgan Kaufmann, (2000).
[7] Ebert-Uphoff, I. Introducing parallel manipulators through
laboratory experiments. IEEE Robotics & Automation Magazine,
10 (3), pp. 13-19, (2003).
[8] Falcone, E., Gockley, R., Porter, E. & Nourbakhsh, I, The personal
rover project, Special Issue on Socially Interactive Robots,
Robotics and Autonomous Systems, (2003).
[9] Gerovich, O., Goldber, R. P., & Donn, I. D. From science projects
to the engineering bench. IEEE Robotics & Automation Magazine,
10 (3), pp. 9-12, (2003).
[10] Hsiu, T., Richards, S., Bhave, A., Perez-Bergquist, A. &
Nourbakhsh, I. Designing a Low-cost, Expressive Educational
Robot. In Proceedings of IROS 2003. Las Vegas, USA, (2003).
[11] Kitts, C. Surf, turf, and above the Earth. IEEE Robotics &
Automation Magazine, 10 (3), pp. 30-36, (2003).
[12] Kumar, D. & Meeden, L. A robot laboratory for teaching artificial
intelligence. In Proc. of 29th SIGCSE Symposium on Computer
Science Education, (1998).
[13] Leinhardt, G. & Crowley, K. Objects of learning, objects of talk:
Changing minds in museums. In S.G. Paris (Ed) Perspectives on
Object-Centered Learning in Museums (pp. 301-324). Mahwah,
NJ: Lawrence Erlbaum Associates (2002).
[14] Leinhardt, G., Crowley, K. & Knutson, K. (Eds.). Learning
Conversations in Museums. Mahwah, NJ: Lawrence Erlbaum
Associates (2002).
[15] Murphy, R. Introduction to AI Robotics. MIT Press, (2000).
[16] Nourbakhsh, I., Crowley, K., Bhave, A., Hamner, E., Hsiu, T.,
Perez-Bergquist, A., Richards, S., & Wilkonson, K. The Robotic
Autonomy Mobile Robotics Course: Robot design, curriculum
design and educational assessment. Autonomous Robotics Journal,
in print. (2004).
[17] Nourbakhsh, I., Hamner, E., Bernstein, D., Crowley, K., Porter, E.,
Hsiu, T., Dunlavey, B., Ayoob, E., Lotter, M., Shelly, S., Parikh,
A., & Clancy, D. The Personal Exploration Rover: The Ground-up
Design, Deployment and Educational Evaluation of an Educational
Robot for Unmediated Informal Learning Sites. Carnegie Mellon
University Technical Report CMU-RI-TR-04-38. August, (2004).
[18] Nourbakhsh, I., Hamner, E., Porter, E., Dunlavey, B., Ayoob, E.,
Hsiu, T., Lotter, M., & Shelly, S. The Design of a Highly Reliable
Robot for Unmediated Museum Interaction. In Proc. International
Conference on Robotics and Automation. IEEE ICRA (2005).
[19] Nourbakhsh, I. When students meet robots. Essay in IEEE
Intelligent Systems and Their Applications, 15(6), p15. (2000).
[20] Nourbakhsh, I. Robotics and education in the classroom and in the
museum: On the study of robots, and robots for study. In
Proceedings Workshop for Personal Robotics for Education. IEEE
ICRA (2000).
[21] Papert, S. & Harel, I. Situating Constructionism, in:
Constructionism, Ablex Publishing Corp., (1991).
[22] Wolz, U. Teaching design and project management with Lego
RCX robots. In Proc. SIGCSE Conference (2000).
... Dès lors, pour faire de cette nouvelle technologie un outil techno-pédagogique efficace pour l'apprentissage, n'est-il pas nécessaire de se pencher attentivement sur les finalités éducatives 5 que l'on souhaite atteindre grâce aux spécificités de cette nouvelle technologie (Resnick & Wilensky, 1993) ? Et enfin, ne faut-il pas trouver des critères d'évaluation appropriés (Nourbakhsh, Hamner, Dunlavey, Bernstein & Crowley, 2005) ? ...
Full-text available
The dyadic relation between technological progress and educational objectives has recently engendered a new field of research at the crossroads of psychology, science education and artificial intelligence: Educational Robotics (ER). This article provides a critical state of the art about ER, its origins and its position in the world of ICTE. It analyzes the educational goals that are achievable according to the status and the specific learning modalities of the different types of robot. Among these different types, a special attention is paid to the robotics kits. Emphasis is given on the educational potential they offer, especially in terms of competences transfer, and on the limits to be overcome in order to meet the challenge of their integration in the school programs.
Bu bölümde FeTeMM yaklaşımının çıkış noktaları, tanımları ve doğasından bahsedilmiştir. Daha sonra teknolojinin farklı tanımları olduğu vurgulanmış ve farklı tanımlar okuyucuya sunulmuştur. Ayrıca, Uluslararası Eğitim Teknolojileri Birliği (The International Society for Technology in Education-ISTE) tarafından ortaya konulan ve öğrencilerin ulaşması istenen kazanımların tanımları özetlenmiştir. Bu teorik kısımdan sonra ülkemizde gerçekleştirilen ve katılımcılara FeTeMM eğitimi sunan toplamda 23 makale teknolojinin kullanım amacı, kullanılan güncel teknolojiler ve çalışmaların yazarlarının alanları bakımından incelenmiştir. Analizler sonucu ülkemizdeki çalışmaların daha çok fen eğitimcileri tarafından gerçekleştirildiği, çalışmalarda teknolojinin daha çok araç olarak kullanıldığı ve en yaygın olarak bilgisayarın kullanıldığı ortaya konulmuştur. Teknolojinin bir disiplin olarak ele alındığı beş çalışmada Uluslararası Eğitim Teknolojileri Birliği tarafından vurgulanan yetkin öğrenen, bilgiyi düzenleyen, yaratıcı tasarımcı, bilişimsel düşünen, yaratıcı iletişimci kazanımlarının kapsandığı ancak dijital vatandaş ve global işbirlikçi kazanımlarının vurgulanmadığı ortaya konulmuştur. Son olarak, uluslararası alan yazında FeTeMM çalışmalarında kullanılan robotik, arttırılmış gerçeklik uygulamaları ve üç boyutlu yazıcılar ve dijital oyunlar hakkında kısaca bilgi sunulmuştur.
While a large number of robotics programs for K-12 students have been developed and deployed in the past twenty years, the effect that these programs have on students' motivations to enter science, technology, engineering, and mathematics (STEM) careers has yet to be fully determined. In order to demonstrate the value of these programs, researchers must make a concerted effort to measure their impact. Based on prior work in the evaluation of educational robotics programs, the authors of this chapter present frequently-utilized evaluation and measurement methods as well as guidelines for selecting these methods based on factors such as a program's duration, size, and maturity. This chapter is intended for use as a reference guide for designing evaluations of K-12 educational robotics programs.
This paper presents the results of a user study in which the impact of sharing decision making in a human-robot team was measured. In the experiments outlined here, a human and robot play a game together in which the robot searches an arena for items, with input from the human, and the human-robot team earns points for finding and correctly identifying the items. The user study reported here involved 60 human subjects. Each subject interacted with two different robots. With one robot, the human acted as a supervisor: the human issued commands and the robot obeyed. With the other robot, the human acted as a collaborator: the human and robot shared decisions and were required to reach agreement about the robot’s actions in the arena before any actions were taken, facilitated using computational argumentation. Objective performance metrics were collected and analyzed for both types of human-robot team, as well subjective feedback from human subjects regarding attitudes toward working with a robot. The objective results showed significant improvement in performance metrics with the human-as-collaborator pairs versus the human-as-supervisor pairs. Subjective results demonstrated significant differences across many subjective measures and indicated a distinct preference for the human-as-collaborator mode. The primary contribution of this work lies in the demonstration and evaluation of a computational argumentation approach to human-robot interaction, particularly in proving the efficacy of this approach over a less autonomous mode of interaction.
Informal science learning aims to improve public understanding of STEM. Free-choice learners can be engaged in a wide range of experiences, ranging from watching entertaining educational videos to actively participating in hands-on projects. Efforts in informal science learning are often gauged by their ability to elicit interaction, to foster learning, and to influence perceptions of STEM fields. This paper presents the installation of a biomimetic robotic fish controlled by an iDevice application at an informal science learning exhibit. Visitors to the exhibit are offered a unique experience that spans engineering and science, in which they can steer the robotic fish, choosing from three modes of control. Visitor engagement is examined through the lens of the Selinda model of visitor learning, while their behavior is examined using an adapted model of Borun's framework for behaviors indicative of learning. The evaluation of the efficacy of the exhibit is assessed through a post-experience survey questionnaire, an analysis of the application usage, and a behavior coding study. Data collected on visitor interactions with the exhibit indicate that free-choice learners value the importance of engineering research, and prefer interactive modes. Further, behavior coding results support the ability of the robotic fish platform to capture the visitors' attention. Findings offer compelling evidence that the exhibit is both highly engaging to visitors and a suitable format for science inquiry.
Robotics has often been proposed as a tool to support informal science education through hands-on activities and learning across fields in science, technology, engineering, and mathematics (STEM). However, instances in which robots have been effectively integrated into informal science learning events are rare, and data on participants' learning and engagement are even harder to find. We present the deployment of a new robotic fish controlled by an iDevice application at a series of informal science learning events. Visitors to the exhibit are offered a unique learning experience and can interact with the robotic fish through a range of control modalities, highlighting the principles of autonomy and remote operations. Data collected on visitor interactions with the application indicate a longer time usage for more interactive control modes with less autonomy. Moreover, survey data garnered after the interaction with the robots suggest that visitors are concerned about the environmental issue of marine pollution and they are eager to learn more. This platform can serve as the basis for systematically exploring the potential of robotics in informal science education by combining refined data collection with interactive control.
While a large number of robotics programs for K-12 students have been developed and deployed in the past twenty years, the effect that these programs have on students' motivations to enter science, technology, engineering, and mathematics (STEM) careers has yet to be fully determined. In order to demonstrate the value of these programs, researchers must make a concerted effort to measure their impact. Based on prior work in the evaluation of educational robotics programs, the authors of this chapter present frequently-utilized evaluation and measurement methods as well as guidelines for selecting these methods based on factors such as a program's duration, size, and maturity. This chapter is intended for use as a reference guide for designing evaluations of K-12 educational robotics programs.
In this article, we present the design, development, and characterization of a biomimetic robotic fish remotely controlled by an iDevice application (app) for use in informal science education. By leveraging robots, biomimicry, and iDevices, we seek to establish an engaging and unique experience for free-choice learners visiting aquariums, zoos, museums, and other public venues. The robotic fish incorporates a three-degree-of-freedom tail along with a combined pitch and buoyancy control system, allowing for high maneuverability in an underwater three-dimensional (3-D) space. The iDevice app implements three modes of control that offer a vividly colored, intuitive, and user-friendly theme to enhance the user experience when controlling the biomimetic robotic fish. In particular, the implemented modes vary in the degree of autonomy of the robotic fish, from fully autonomous to remotely controlled. A series of tests are conducted to assess the performance of the robotic fish and the interactive control modes. Finally, a usability study on elementary school students is performed to learn about students' perception of the platform and the various control modes.
Conference Paper
Interactive robotics in formal and informal settings alike has been shown to effectively excite and educate learners at every level. In this second of two papers, we present the educational application of recently-developed biomimetic robotic fish for K-12 learning at the New York Aquarium focused on underwater robotics and marine science. We narrate the development, organization, and execution of an outreach program designed around these robotic fish to pique K-12 students’ interest in science, technology, engineering, and mathematics (STEM) and to attract them toward engineering careers. The activity offers an authentic engineering experience through bioinspired modification of the swimming robots informed by observation of the aquarium’s inhabitants. Student survey responses indicate the success of the activity in influencing the students’ perception of engineering. More specifically, the students showed an increased interest in STEM fields and found engineering to be a more accessible and exciting discipline after the activity.
Full-text available
Current accounts of the development of scientific reasoning focus on individual children's ability to coordinate the collection and evaluation of evidence with the creation of theories to explain the evidence. This observational study of parent–child interactions in a children's museum demonstrated that parents shape and support children's scientific thinking in everyday, nonobligatory activity. When children engaged an exhibit with parents, their exploration of evidence was observed to be longer, broader, and more focused on relevant comparisons than children who engaged the exhibit without their parents. Parents were observed to talk to children about how to select and encode appropriate evidence and how to make direct comparisons between the most informative kinds of evidence. Parents also sometimes assumed the role of explainer by casting children's experience in causal terms, connecting the experience to prior knowledge, or introducing abstract principles. We discuss these findings with respect to two dimensions of children's scientific thinking: developments in evidence collection and developments in theory construction. © 2001 John Wiley & Sons, Inc. Sci Ed85:712–732, 2001.
Full-text available
Robotic Autonomy is a seven-week, hands-on introduction to robotics designed for high school students. The course presents a broad survey of robotics, beginning with mechanism and electronics and ending with robot behavior, navigation and remote teleoperation. During the summer of 2002, Robotic Autonomy was taught to twenty eight students at Carnegie Mellon West in cooperation with NASA/Ames (Moffett Field, CA). The educational robot and course curriculum were the result of a ground-up design effort chartered to develop an effective and low-cost robot for secondary level education and home use. Cooperation between Carnegie Mellon's Robotics Institute, Gogoco, LLC. and Acroname Inc. yielded notable innovations including a fast-build robot construction kit, indoor/outdoor terrainability, CMOS vision-centered sensing, back-EMF motor speed control and a Java-based robot programming interface. In conjunction with robot and curriculum design, the authors at the Robotics Institute and the University of Pittsburgh's Learning Research and Development Center planned a methodology for evaluating the educational efficacy of Robotic Autonomy, implementing both formative and summative evaluations of progress as well as an in-depth, one week ethnography to identify micro-genetic mechanisms of learning that would inform the broader evaluation. This article describes the robot and curriculum design processes and then the educational analysis methodology and statistically significant results, demonstrating the positive impact of Robotic Autonomy on student learning well beyond the boundaries of specific technical concepts in robotics.
In this study, 178 groups of visitors were interviewed and recorded during their visits to museums. Three clusters of elements were shown to influence learning: the identity of the visitors, their response to the learning environment, and their explanatory engagement during the visit. A structural equation model using these variables fit well. Further examination revealed that not all conversational behavior was supportive of learning; some actions, such as making frequent personal connections, were detrimental to learning; additionally, silent contemplation was modestly associated with learning. This paper discusses these findings through the experiences of four couples whose outcome measures placed them at the extreme high or low end of the learning distribution.