Publication Date

8-2025

Date of Final Oral Examination (Defense)

4-25-2025

Type of Culminating Activity

Thesis

Degree Title

Master of Science in Computer Science

Department

Computer Science

Supervisory Committee Chair

Casey Kennington, Ph.D.

Supervisory Committee Member

Edoardo Serra, Ph.D.

Supervisory Committee Member

Bogdan Dit, Ph.D.

Abstract

Recent advancements in language modeling have improved robotic emotion expressiveness, yet several challenges remain. Many existing robotic expression models rely on fixed rules and static frameworks, which limit their ability to capture the dynamic nature of emotional expression. Additionally, these systems often struggle to balance emotional accuracy with generating novel and varied behaviors. These limitations underscore the need for method capable of delivering both emotionally accurate and diverse robot behaviors.

In this thesis, we addresses these challenges by introducing a new framework for expressive behavior generation in robots. We develop a generative model that generates robot behavior sequences that align with specific emotional inputs. To further improve its behavior generation, we introduced an adversarial framework-inspired training approach that incorporates feedback from both the emotion classifier and a novelty function. This dual feedback mechanism helps the model improve not only its emotional accuracy but also its ability to generate diverse and non-repetitive behaviors. Experimental findings from human evaluations confirm that the model effectively generates behaviors that are both emotionally expressive and clearly distinct.

DOI

10.18122/td.2392.boisestate

Included in

Robotics Commons

Share

COinS