Guiding Users Interactively through Difficult Environments is the answer of Universidad Complutense of Madrid for the first GIVE challenge. GUIDE is pretend to be an architecture to develop custom virtual guides focusing on the user capabilities.
A virtual guide based on GUIDE senses the environment just like the user does and is able to commit instructions in a more human fashion than other guides. It manages a set of agents ordered of multiple tasks such as driving the user, disambiguating actions, taking care of the user behaviour, warning or informing him...
At present time, GUIDE theorical architecture is complete but the main ideas are actually implemented between the system presented in GIVE challenge and the current development version, GUIDE2. We expected to complete GUIDE2 in the near future in order to present it to the second installment of GIVE challenge.
Try it yourself (coming soon!)
Main contributors
- Daniel Dionne (main developer)
- Salvador de la Puente (main developer)
- Carlos León (director)
- Raquel Hervás (director)
- Pablo Gervás (director)
Related publications
- A Model for Human Readable Instruction Generation Using Level-Based Discourse Planning and Dynamic Inference of Attributes Disambiguation (Conference Paper, 2009) wrote by Dionne D, de la Puente S, León C, Hervás R, Gervás P and published at 12th European Workshop on Natural Language Generation.
[See its bibliographic reference] - Constructing World Abstractions for Natural Language in Virtual 3D Environments (Conference Paper, 2009) wrote by 2nd International Symposium on Intelligent Interactive Multimedia Systems and Services.
[See its bibliographic reference]
Other related documents
- Development details contains a short story about the development process talking about the main ideas and challenges we had to face.
- Download the poster in printable format.