Research on the Application of Intelligent Sensor Data Analysis in Optimizing Martial Arts Training Movement Techniques

The martial arts training methods are essential for enhancing performance, minimizing injury risks, and refining movement precision. Integrating intelligent sensor expertise into sports training has opened possibilities for real-time analysis and feedback. Sensors like gyroscopes, pressure, and accelerometers can suggest valuable approaches to athletes’ biomechanics, allowing for a deeper understanding of their movements and areas for development. The research aims to explore the application of intelligent sensor data investigation in optimizing martial arts training movement methods. Data were collected using wearable sensors, including accelerometers, pressure sensors, and gyroscopes placed on the body parts of martial artists throughout training sessions. Each participant underwent a series of prohibited training movements similar to punches, kicks, and footwork. To record sensor data in real-time during these movements, capturing biomechanical parameters like angular velocity, force, and acceleration. The research proposes a novel intelligent honey badger optimization-tuned dynamic decision tree (IHBO-DDT) model for more accurate, competent modeling of martial arts movements, optimizing performance, and predicting injury risk by adapting to intricate, dynamic training conditions with minimal error. Findings demonstrate that the IHBO-DDT approach considerably enhances the accuracy of movement analysis, identifying incorrect techniques and inefficiencies. Regarding optimization, the IHBO-DDT technique outperforms ST-HKA and SVR, achieving an accuracy of 95.5%, F1-score of 92.4%, and AUC of 0.805 while decreasing execution time by 10 min. The proposed model for martial arts training provides a promising data-driven strategy to optimize performance and forecast injury risks.