Abstract
This paper presents the MADbot architecture which enables a motivated autonomous agent to generate goals in response to changes in its underlying drives or motivations, while it is both planning and executing. We use a Mars rover domain to illustrate the workings of this architecture and describe two ways of modelling motivations and goal generation: the first decouples the model of motivations and goal generation from the workings of an AI planner; the second models motivations and goal generation implicitly in the planner’s domain model. We discuss the advantages and disadvantages of each approach and offer ideas for further work.
| Original language | English |
|---|---|
| Pages | 39-46 |
| Number of pages | 8 |
| Publication status | Published - 15 Dec 2006 |
| Event | 25th Workshop of the UK Planning and Scheduling Special Interest Group - Nottingham, United Kingdom Duration: 14 Dec 2006 → 15 Dec 2006 |
Conference
| Conference | 25th Workshop of the UK Planning and Scheduling Special Interest Group |
|---|---|
| Abbreviated title | PlanSIG 2006 |
| Country/Territory | United Kingdom |
| City | Nottingham |
| Period | 14/12/06 → 15/12/06 |
Keywords
- MADbot
- modelling motivations
- goal generation
Fingerprint
Dive into the research topics of 'Motivations for MADbot: a Motivated and Goal Directed Robot'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver