From sensor data to coaching in alpine skiing – A software design to facilitate immediate feedback in sports

Abstract

Thanks to wearable sensor technologies, it has become feasible to quantify human kinematics cheaply and comprehensively during sports. However, it is often left to the user to infer any qualitative information from the data, leaving them confused about their performance and what actions to take next. This paper presents a high-level process to transform sensor data into immediate expert feedback in the form of coaching instructions. Individual aspects of process and software design are discussed based on an example implementation for Alpine skiing. In detail, this paper aims to (1) describe the transformation from raw sensor data into coaching instructions from a software engineering and data-centric perspective; (2) propose a high-level software design for coaching applications in sports that facilitates historical as well as immediate data analytics; (3) decompose the task of developing coaching applications into independent, manageable research subtasks; and (4) show software engineers which data structures and interactions to implement.