Wind-Normalized Southeast Asian 100-m Sprint Performances and a Conceptual Framework for AI-Powered Exoskeletons

This study has two distinct components. First, an empirical secondary analysis applies a quadratic wind-normalization model to 20 recorded Southeast Asian (SEA) men's 100-m performances for which official times and wind readings were available in the compiled dataset. The mean official time was 10.399 ± 0.177 s and the mean model-adjusted time was 10.379 ± 0.185 s. The adjustment changes the relative ordering of some performances, illustrating how wind conditions can affect historical comparisons. Second, the paper presents a conceptual framework for an AI-powered lower-limb exoskeleton using inertial sensing, temporal-convolutional and recurrent neural-network components, athlete-specific digital twins, phase-specific assistance, and independent safety supervision. No exoskeleton intervention or athlete trial was conducted in this study; therefore, its proposed effects on sprint kinematics, performance, neuromuscular function, and injury risk remain hypotheses requiring controlled validation. The combined contribution is a wind-adjusted regional performance analysis and a technology proposal motivated by the training and performance questions that such comparisons raise.