Prediction of on-stride walking for pregnant women

Pregnant women's size, shape, and weight changes have significant effects on their walking stability. Traditionally, experiments are used to study the effects of subjects, but it is time consuming and expensive. This paper presents an optimization-based pregnant women walking simulation with one-stride formulation. The pregnant woman's model with 55 degrees of freedom (DOFs) is used, including 6 global DOF's and 49 human body DOF's. The dynamic equations of motion are based on the recursive dynamics. Without the constraint of symmetry of the human body between two steps within one walking cycle, the study is based on biomechanical, human kinematic, and dynamic properties to perform the one-stride simulation, which represent the holonomic and non-holonomic constraints in walking simulation. This forms a nonlinear optimization problem. The summation of all joint actuator torques squared within one stride is the cost function. Nine determinant DOF's are used to analyze the kinematics and three for dynamics. Three cases (non-pregnancy, 6 month, and 9 month pregnancy) are adopted for the test and investigation. The simulation results show that during the course of pregnancy, pregnant women's bodies dynamic and kinematic properties change and thus affect their walking and stability.