Skip to main content
Thesis defences

PhD Oral Exam - Rukiye Kirgil Budakli, Information and Systems Engineering

Robot-supervised intelligent workload reallocation based on stress-aware human performance monitoring in human-robot teams


Date & time
Monday, August 4, 2025
1 p.m. – 4 p.m.
Cost

This event is free

Organization

School of Graduate Studies

Contact

Dolly Grewal

Accessible location

Yes

When studying for a doctoral degree (PhD), candidates submit a thesis that provides a critical review of the current state of knowledge of the thesis subject as well as the student’s own contributions to the subject. The distinguishing criterion of doctoral graduate research is a significant and original contribution to knowledge.

Once accepted, the candidate presents the thesis orally. This oral exam is open to the public.

Abstract

The integration of humans and artificial intelligence-based robotic systems in collaborative environments is transforming the landscape of teamwork across various domains. These human–robot teams, which include both physically embodied robots and intelligent virtual agents, require careful coordination to ensure effective task performance. A critical component of such collaboration is the dynamic allocation of workload, which must account for the distinct characteristics and interaction modes of human and robotic agents. Human performance, influenced by emotional and physiological factors such as stress, contrasts sharply with the algorithmic and cognitive nature of robotic behavior. This disparity underscores the need for adaptive task allocation strategies that prioritize human well-being while maximizing overall team efficiency.

This research investigates a robot-supervised, stress-aware workload allocation framework that continuously monitors human stress levels and reallocates tasks in real time to maintain optimal human performance. Anchored in the Yerkes-Dodson law, the proposed approach assumes that human efficiency peaks within a moderate stress range and deteriorates when stress exceeds or falls below this threshold. By leveraging recent advancements in wearable technology and affective computing, this study explores multiple physiological (e.g., EEG, f-NIRS, ECG, EDA, EOG) and behavioral (e.g., facial expressions, speech, eye movement) indicators to quantify and track human stress. The research further considers contextual factors such as task complexity, time of day, and individual differences in skills and knowledge, all of which can impact stress and performance.

The core objective is to develop a real-time decision-making mechanism by which robots can autonomously determine when and how to intervene in task distribution based on fluctuations in human stress. A structured research framework, comprising a main research question and ten sub-questions, guides the development of this model, addressing key aspects such as task zones, interaction modes, communication channels, performance metrics, and stress quantification techniques. Several foundational assumptions support the model, including the ability to distinguish task domains between humans and robots, the quantifiability of human stress, and the application of control charts for detecting deviations in performance over time.

Ultimately, this thesis contributes a novel, stress-sensitive task reallocation algorithm designed to optimize collaborative performance in human–robot teams. By enabling robots to detect stress-induced performance changes and adapt task assignments accordingly, the proposed system aims to enhance both the resilience and productivity of hybrid teams operating in dynamic environments.

Back to top

© Concordia University