of their robot, while allowing them the flexibility to
customize the platform’s hardware later if they wish.
The second largest challenge was coordination
across the different teams. The notion of the explicit
focus groups also emerged throughout the first offering of this course, and was made a key component of
the course thereafter. The use of focus groups, coupled with the use of the shared hardware platform,
allows students to easily share functional developments across teams, placing the emphasis on integration rather than the development of individual
capabilities. One of the products required of each
team and focus group is a website with detailed
instructions and an associated code repository that
other teams, and even future students in the course,
can build upon, lending longevity to the projects.
The use of an agile development methodology (such
as Scrum) along with online project management
software (for example, Trello or Pivotal Tracker) is
essential to keep all of these different teams and focus
groups coordinated, and to ensure continual progress
toward completing the projects.
The types of service tasks explored in the project
were partly inspired by RoboCup@Home (van Beek et
al. 2015) — an international competition where
robots perform relatively simple domestic or commercial service tasks in real environments with all
4 This project-driven course could
easily lead to fielding a RoboCup@Home team, providing students with the opportunity to continue
developing service robots.
Finally, although this service robotics course was
designed at the graduate level, it could be adapted to
the advanced undergraduate level. With some refinement and improved documentation, the robot hardware and software infrastructure developed over the
past two semesters could be adapted to provide scaffolding for undergraduate projects in service robotics, either as part of a course in service robotics or as
a focused senior capstone experience. Instead of
focusing solely on current research papers, the syllabus could be revised to include lectures on service
robotics, ROS, various AI and robotics techniques,
and architectures for integrating those techniques
together. However, graduate students who have taken the course typically find the service robot project
intimidating and challenging, and so special care
would need to be taken to make the course and project accessible to undergraduates.
Teaching Integrated AI at
the Undergraduate Level Through
Socially Aware Projects
Although the service robotics project in its current
form may be better suited to graduate study, many of
the same ideas can be used to teach integrated AI at
the undergraduate level through project-driven
courses. As an example, this section describes the
undergraduate special topics course called Computa-
tional Sustainability and Assistive Computing (figure
6) that was taught at Bryn Mawr College in fall 2010.5
The course focused on the use of computational
Figure 5. Heimeier’s Catechism.
What are you trying to do? Articulate your objectives using
absolutely no jargon.
How is it done today, and what are the limits of current
What is new in your approach and why do you think it will
Who cares? If you succeed, what difference will it make?
What are the risks?
What are the midterm and ;nal “exams” to check for
How much will it cost? How long will it take? (optional for
Heilmeier’s Catechism is a set of questions that should be
addressed in any research proposal. These questions are
credited to George H. Heilmeier, former director of the
Defense Advanced Research Projects Agency (DARPA),
and former president and CEO of Bellcore (DARPA, 2016).
Figure 6. Computational Sustainability
and Assistive Computing Course Description
Explore the use of computers and computational methods
for positive change, examining both broader impacts on
societal development and environmental sustainability,
and narrower improvements to individual lives through
assistive technologies. We will cover a variety of interdisciplinary topics, including computational allocation of
natural resources, monitoring societal-environmental interactions and impacts, ecological modeling, green
computing, assistive technologies for people with dis-abilities, telemedicine, and computers in the developing