Development and Validation of a Gyroscope-Based Turn Detection Algorithm for Alpine Skiing in the Field.

Area of Science:

• Sports Science
• Biomechanics
• Engineering

Background:

• Accurate detection of ski turn switches is crucial for performance analysis.
• Previous methods using inertial measurement units (IMU) showed promise in lab settings but lacked field validation.
• Real-world skiing introduces variables like slope, speed, and turn style that complicate data collection.

Purpose of the Study:

• To develop and validate an IMU-based system for accurately detecting alpine ski turns during field conditions.
• To assess the system's performance across various turn styles, including carving, drifting, and snowplow.
• To identify complete ski runs using the developed methodology.

Main Methods:

• Utilized inertial measurement units (IMU) to collect kinematic data during on-snow skiing.
• Developed and refined a turn detection algorithm to process IMU data.
• Evaluated algorithm performance using metrics such as ratio, precision, and recall for different turn types.
• Tested the system across diverse skiing conditions and turn styles (long carving, short carving, drifted, snowplow).

Main Results:

• The validated system demonstrated high accuracy and validity in detecting ski runs and carved turns (long and short).
• Performance for drifted turns showed high precision but missed some actual turns.
• Snowplow turns exhibited lower accuracy, indicating a need for algorithm improvement.
• Ratio and precision values varied significantly across turn styles, with carved turns performing best.

Conclusions:

• The developed IMU-based system is accurate and valid for detecting ski runs and carved turns in field conditions.
• Further algorithm refinement is necessary to reliably detect drifted and snowplow turns.
• The system shows potential for objective ski performance analysis, especially for advanced turning techniques.