More about HKUST
Computational Approaches to Cognition Alignment in Human-Robot Teaming for Improved Collaborative Performance: A Survey
PhD Qualifying Examination Title: "Computational Approaches to Cognition Alignment in Human-Robot Teaming for Improved Collaborative Performance: A Survey" by Miss Ziqi PAN Abstract: Robots are increasingly taking on collaborative roles across a diverse range of human activities. This emerging partnership, defined as Human-Robot Teaming (HRT), presents more challenges to teamwork beyond human teams. Research has modeled approaches to enhance Human-Robot Teaming (HRT) performance by distributing optimized factors within the Input-Mediator-Outcome (I-M-O) framework. This framework identifies team and task traits (I), applies methods to leverage these traits (M), and aims for better collaborative performance (O). Cognition alignment, which focuses on leveraging humans' and robots' cognitive intelligence, is remarkable among the various existing strategies. In this survey, we focus on aligning cognition between humans and robots to enhance the collaborative performance of HRT. We first review the literature, identify key metrics of collaborative performance, and categorize them as human recognition, performance and load, as-a-team performance, and within-team dynamics. Then, we summarize and analyze recent advancements in computational approaches to cognition alignment based on the I and Ms. We first examine the strengths and weaknesses of different approaches and then discuss their contribution regarding the aforementioned types of collaborative performance metrics in HRT. Finally, based on the findings and discussion, we propose emerging trends and future directions for cognition alignment to enhance collaborative performance in HRT. Date: Monday, 10 March 2025 Time: 3:00pm - 5:00pm Venue: Room 4472 Lifts 25/26 Committee Members: Dr. Xiaojuan Ma (Supervisor) Prof. Qiong Luo (Chairperson) Dr. Tristan Braud