.png){kind=icon}
Return
Programming an expressive autonomous agent
.png){kind=icon}
PRE
AI
DOI：10.1016/j.eswa.2015.08.037.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Emotions are the integral part of human cognitive processes. To make an artificial agent more human-like, agent programming languages should be able to model the emotion based artificial agent. To address the need of incorporation of emotions in agent programming languages, recently, extension of 2APL with emotions is proposed by the researchers. However, they were unable to program complex emotion dynamics, due to integration complexities associated with advanced emotive models. They have modified 2APL by designing E-rules, coping rules, and an interface-Affect Engine. The Affect Engine establishes a link between the modified 2APL and a rather simpler emotion model, ALMA. Here, we propose the programming constructs for integrating an advanced yet rule based emotion model, EMIA, in line with the 2APL. The fusion of both has been carried out by redefining the syntax, semantics and deliberation cycle of the 2APL. These redefinitions lead to the required transformation in the dialect cycle; from the sense-interpret-act to percept-appraisal-elicitation-act. The proposed novel programming constructs and pragmatics enable flexible and adaptive emotional behavioral modeling for an expressive autonomous agent with domain independent emotion elicitation, emotion regulation and emotion transition processes. The simulation results show high believability in the emotions expressed by the agent while responding to the real life scenarios. (C) 2015 Elsevier Ltd. All rights reserved.
Keywords:
Autonomous agent
Agent programming language
Emotion engine
Emotion modeling
2APL
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
7.5
Papers:
2.9W
Citations:
10.2W
Organization
No organization information available
Cited Papers
.png){kind=icon}
ShareSaveShareSaveShareSaveShareSaveShareSave
ShareSave
View more