Abstract:
Foot landing is crucial in daily activities and sports, providing balance, stability, and energy efficiency while reducing injury risk. Proper landing distributes impact forces evenly, preventing overuse injuries. In racquet-based sports like badminton, proper foot landing and racquet handling are believed to be key factors that affect players' performance. During athletic activities or training, players can make mistakes or movements that are potentially harmful if not addressed. Therefore, performance analysis is crucial in these situations. While numerous studies have examined the biomechanical characteristics of the upper limb, including various stroke and posture analyses at different levels of gameplay, research on the lower limb has been relatively scarce in racquet-based sports. The main objective of this study was to develop a wearable motion sensor system to correctly classify foot landing, specifically distinguishing between heel and toe landing during a game of badminton, to discern the high impact on the lower limb, which is susceptible to injury. For the classification of foot landing, we developed a machine-learning algorithm to evaluate its performance. Eight healthy participants (age: 21.4±1.5 years) were analyzed. Experimental results indicate that our trained algorithm achieves an accuracy of 97.53% for foot landing classification.