Session Multi-Level Project Work A Study in Collaboration: Mats Daniels and Lars Asplund
Session Multi-Level Project Work A Study in Collaboration: Mats Daniels and Lars Asplund
Session Multi-Level Project Work A Study in Collaboration: Mats Daniels and Lars Asplund
the corresponding modified Hue signal. The purpose of the • Measuring of distance
modification is to be able to also capture very light [white] • Camera subsystem
and very dark [black] signals. The modified Hue signal is
then used in another table where different features, up to • Internal communication using the CAN-bus
eight, are coded. This feature signal is sent via a fast • External communication using Blue Tooth
connection to a digital signal processor. The streams • Measurements of angles between robots in the own
between cameras and signal processors are controlled with team
switches connected to the video processors and can for
instance be directed to use a single camera. • Kick-subsystem
Motion is done with two wheels. The wheels can be • Motor and tachometer
positioned in all possible directions. The turning is
• Collision avoidance
controlled via stepping motors and the driving of the wheel
is by a DC motor. In an early draft an external company was The Mechanical engineers
approached to construct these wheels, but it couldn’t be done
due to lack of resources. The mechanical engineering students are the ones with the
The ball is handled with a “catcher”, where the ball can most concrete task and also the least time. They do their
be pushed ahead of the robot. There is also a “kick” contribution as part of a three credit course. The specific
equipment. This is pneumatic and driven by gas controlled tasks for them has been
by a magnetic vent. • The kicker
Sponsors • The wheels
• The chassis
The budget for this project is based on the normal course
budget and sponsoring from industry. Six companies are Time line
each sponsoring a robot. In the game only four can
participate, but having six is important in the development The electrical and the mechanical engineers only work with
process. Practice can for instance be done between two the project for the first half of the semester, whereas the IT
teams of three robots. There are also a number of other engineers work throughout the whole semester. We do
sponsors that have donated different components. anticipate that several of the Electrical and Mechanical
engineers will continue in the project by doing thesis work
THE STUDENT GROUPS during the second half of the semester.
The IT engineers have set up three major deadlines. The
The courses officially start mid-January and run till end of first is when the other groups leave the project in mid
May and end of March respectively. All classes have March, and the second is late April and the third and final at
roughly 30 students and both have started preparations for the end of the semester in late May. Much of the
the project during the fall semester. The electrical engineers preparations started already in the fall semester, when
have been using components that will be part of the robots in feasibility studies were conducted.
a course covering real-time and distributed systems and the
information technology students have formed subgroups INTEGRATION BENEFITS
with different responsibilities and have started to gather
information and also meet with their corresponding It has been educational and interesting to follow how the
subgroups among the electrical engineers. The mechanical different student groups approached each other. There were
engineers came in relatively late in the planning, but have not much prior knowledge about the other groups and
proven to be very useful. especially their respective competence. One issue has been
to settle on the level of the assignment. It is interesting to
The IT engineers note that even if there is a clear “us” and “them”
The IT engineers are divided into two groups. One group environment there doesn’t seem to be a hierarchy.
will handle sensor fusion, i.e. to gather a uniform view of the There is a mutual respect between all three groups. The
current situation, and the other will do the AI-part, i.e. to relatively high difference between the groups is probably
play the game given the unified view from the current contributing to this respect, since all groups have a need for
situation. each other. The spread has also made it possible to run a
project with a reality touch that goes beyond what we have
The Electrical engineers previously been able to do. The project addresses AI issues
The electrical engineering students are split into six groups that rely on underlying functions taking in dealing with the
and each will use a small sized management tool for information from the environment, which is taken in by
specifying the requirements, given by the information sensors, and all being made possible by mechanical design.
technology students. The six groups will deal with: There is interdependency and a need for understandable
GRADING ISSUES
The work in this project is for the three groups evaluated
against a Pass/Fail scale. This makes handling courses like
this easier, since they usually contain a wide variety of
assignments, which are difficult to compare if more
differentiated grading is done on an individual basis.
CONCLUSIONS
The project is extremely fun to work in and have a tendency
to engage the students at a probably too high time quota. The
students are doing this at full time, i.e. no other courses in
parallel, but consideration on the time investment will be
taken for future instances. There is a demand for equipment
that is different than our normal situation and it has been
slightly problematic to get our hands on suitable equipment.
Early observations indicate that having a concrete and
exciting goal with firm deadlines is successful. The fact that
the project has a touch of “put a man on the moon” type and
that it is run in a realistic setting with a wide range of
competence required also seem to be of high value for the
spirit in which work is done. It is our belief that the setting
promotes professional skills and exposes the students to a
situation where learning to learn is a natural component.
Other observed benefits are training in communication
between interdisciplinary experts, as well as being a positive
factor in forming, or rather adding to, an identity for the
students groups.
REFERENCES
[1] Daniels, M. & Asplund, L., Full Scale Industrial Project Work, a one
semester course, Proc. IEEE Frontiers in Education, San Juan, 1999.
[2] Mosiman, S. & Hiemcke, C., Interdisciplinary Capstone Group
Project Designing Autonomous Race Vehicles, Proc. ACM SIGCSE
Technical Symposium on Computer Science Education, Austin, 2000.