From Data to Coaching: A Survey of AI-Mediated Feedback Loops in Adventure Sports Training

PhD Qualifying Examination


Title: "From Data to Coaching: A Survey of AI-Mediated Feedback Loops in
Adventure Sports Training"

by

Mr. Yujia HE


Abstract:

AI-mediated coaching systems increasingly transform sensor data, video streams,
and biomechanical signals into performance feedback intended to guide athletes
beyond traditional coach–athlete settings. Within Sports HCI, prior work has
explored diverse feedback modalities—including visual analytics, augmented and
mixed reality, multimodal sensing, and machine learning—yet much of this work
implicitly centers on closed or mechanically constrained tasks. Less attention
has been paid to how such systems operationalise coaching in environmentally
open sports, where performance unfolds in dynamic, non-repeatable conditions
that demand continuous perception–action coupling and adaptive decision-making.

This PQE presents a critical survey of AI-mediated coaching systems, with
particular attention to environmentally open, self-paced sports (e.g.,
climbing, diving, and skiing) and comparative contrasts with closed-skill
domains. The review introduces a learning-cycle taxonomy grounded in
established coaching and feedback theory to analyse how systems sense,
interpret, and intervene within athlete learning processes. Rather than
categorising systems by technology alone, this framework examines how
AI-mediated tools structure feedback loops over time and how they position
athletes in relation to expertise, automation, and adaptation.

By interrogating prevailing assumptions about optimal form, objectivity, and
short-term correction, this work identifies conceptual and methodological gaps
in how current systems support sense-making, contextual judgment, and
longitudinal skill development in everyday sports practice. The analysis
highlights challenges and opportunities for designing AI-mediated coaching
technologies that better account for environmental variability, embodied
exploration, and situated learning.


Date:                   Friday, 27 February 2026

Time:                   3:00pm - 5:00pm

Venue:                  Room 3494
                        Lift 25/26

Committee Members:      Prof. Huamin Qu (Supervisor)
                        Prof. Qian Zhang (Chairperson)
                        Dr. Arpit Narechania