Project Details
- Project OverviewSEGAL assisted the trans-Atlantic rowers of OAR Northwest through the development of an adaptive scheduling system that will help them keep on task as they become increasingly mentally and physically fatigued during their voyage. Our research on this project includes runtime requirements elicitation in unobservable environments.
- SEGAL Members InvolvedAlessia Knauss, Angela Rook, Daniela Damian
- Research PartnersOar Northwest, Centre for Sleep and Human Performance
Project Description
On January 23rd 2013, four world-class athletes from OAR Northwest, a long-distance rowing organization partnered with oceanographic and athletic research groups, launched their modified rowboat from Dakar, Africa. Their goal was to cross the Atlantic Ocean using only their own manpower, twenty-four hours per day in shifts of two rowers at a time, and arrive in Miami, Florida to set a Guiness World Record for being the first to cross unassisted from Africa to North America in a rowboat.
This mentally and physically intensive task of crossing five time zones was supported by the SEGAL lab through developing a location-sensitive, adaptive scheduling system for two major purposes:
- to assist the rowers in cognitive offloading as they become increasingly physically and mentally fatigued as the journey progresses by alerting them when to complete tasks ranging from sleep/row shifts, to taking scientific readings, to brushing their teeth
- to reduce physical and mental stress on the rowers by automatically adapting the system’s ‘boat time’ by tiny increments based on longitudinal position instead of large hourly timezone changes
Through post-trip analysis of the contextual data the rowers collected during their journey, we hope to be able to gain valuable insight into runtime requirements elicitation, particularly in unobservable environments.
Development of Task on Time Executive Manager (ToTEM)
Alessia Knauss, Jason Cummer (also involved in the ToTEM stage of this project), and Angela Rook with prospective hardware for the ToTEM prototype.
ToTEM, a scheduling system and context-adaptive time adjuster, was developed as part of a Computer-Supported Cooperative Work course in the Department of Computer Science at UVic. This course was taught by Associate Professor and SEGAL director Daniela Damian.
Our research investigates how to properly design system adaptations by considering the unobservable operational environmental contexts. However due to the nature of the operational setting, we were not able to observe the users during development. The rowers collected a range of data from their voyage which will help us understand how the system should adapt to certain contexts, and will be used to improve future versions of the system.
This investigation is in partial fulfillment of Alessia Knauss‘ PhD requirements.
Identifying Contexts of Performance Anomalies using Data Mining
For a preliminary study, Angela Rook (using data mining techniques) and Alessia Knauss investigated a subset of data from a previous OAR Northwest row circumnavigating Vancouver Island in April, 2012 (see video above). Currently, we are further investigating elicitation techniques using this data and the results will inform our analysis of the trans-Atlantic Dakar to Miami voyage.
- The National
Adam Kreek reflects on his transatlantic journey - National Post
Adam Kreek Recalls the Moment Rowboat Capsized in the Middle of the Atlantic Ocean - Matt Lauer and the Today Show
Rowers who attempted record are rescued at sea - ESPN
A Challenging Row – From Africa to the U.S. - The Globe and Mail
Former Olympian Adam Kreek readies for a different kind of challenge - Times Colonist
Olympian rower leads ocean odyssey - The Ring
Rowing Across the Atlantic with UVic’s help
Angela Rook, Alessia Knauss, Daniela Damian, and Alex Thomo. (2014.) A Case Study of Applying Data Mining to Sensor Data for Contextual Requirements Analysis. In Proceedings of International Workshop on Artificial Intelligence for Requirements Engineering (AIRE ’14). Karlskrona, Sweden. (PDF)
Angela Rook, Alessia Knauss, Daniela Damian, Hausi A. Muller, and Alex Thomo. Integrating Data Mining into Feedback Loops for Predictive Context Adaptation. Technical Report DCS-349-IR, University of Victoria, Victoria BC, August 2013 (PDF).
Alessia Knauss. On the Usage of Context for Requirements Elicitation: End-User Involvement in IT-Ecosystems. Proceedings of 20th International Requirements Engineering Conference (RE ’12), pages 345-348, Chicago, USA, 2012. Doctoral Symposium.
Olesia Brill*, and Eric Knauss. Structured and Unobtrusive Observation of Anonymous Users and their Context for Requirements Elicitation. Proceedings of 19th International Requirements Engineering Conference (RE ’11), Trento, Italy, 2011 (* Please note that Brill is Alessia Knauss’ maiden name)
Comments are closed.