Dynamic Simulation of Human Breaststroke Based on Inertial Sensor Measurements and Multi-rigid-body Model

Swimming is a sport that involves complex limb and body movements, making it challenging to measure and simulate human swimming. This research successfully implemented the entire process of swimming analysis, from measuring human swimming motion to dynamic simulations. The acquisition of human swimming signals is the basis of this study. Inertial sensors are inexpensive and easy to deploy and therefore were used to measure human motion signals. In this study, measurements from inertial measurement units (IMU) were compared with an optical motion capture system to verify the validity, reliability, and accuracy of the measurements. The obtained human motion signals were then input into the multi-rigid body model of the human body for dynamic simulation. The dynamic model of the human body comprises 18 rigid bodies, whose shapes are determined according to the geometric characteristics of the human subject. Different resistances to the rigid bodies, including passive and active fluid resistances, are also taken into account. The simulation results were highly reliable, providing valuable insights into the interaction between human swimming and water current. The experimental findings highlight the potential of IMUs for effectively measuring human motion, particularly in human breaststroke swimming.