Simulating surfing with optimal control: Sensorfusion for biomechanical analysis

Currently, there is no established biomechanical model for surfing. Especially, musculoskeletal simulations can provide valuable insights for athletes to enhance performance and prevent injuries using quantitative data such as joint angles, joint moments, and muscle forces. The dynamic nature of surfing makes it difficult to assess biomechanics in a laboratory setting. Although inertial measurement units (IMUs) offer a potential solution, relying solely on IMU data in three-dimensional musculoskeletal models presents challenges. We hypothesize a multi-modal approach combining IMU data with deep learning-based human pose estimation to simulate surfing movements. We collected data from seven experienced surfers on a river wave, providing a semi-controlled environment. We placed IMUs on the participants’ bodies, and positioned cameras around the1wave. We defined an optimal control problem to drive three-dimensional biomechanical simulations. In addition, we used an adapted contact model to represent the interaction between feet, surfboard, and water surface. Consequently, we compared the kinematics and kinetics of the front and rear legs, and analyzed muscle forces, to demonstrate the potential of our approach. The study successfully minimized tracking errors and established the first biomechanical model for surfing. Differences between rear and front leg joint moments and angles of up to 47% were found, with muscle forces being higher in the front leg’s muscles, suggesting that the weight distribution in surfing comes mainly from the front leg muscles, while the higher rear leg moments come from stabilization. This showcases the high potential for further biomechanical analysis in surfing performance analysis using our approach.