Effects of IMU placement strategies on the accuracy of hip joint angle estimation

Inertial measurement units (IMUs) provide a portable and low-cost solution for motion monitoring, but their estimation accuracy is strongly influenced by sensor placement. This study designs five IMU placement strategies with different numbers and positions of sensors, based on data from sixteen healthy participants performing six typical hip joint movements. Hip joint angles in the sagittal, frontal, and transverse planes are estimated using artificial neural network (ANN). The results show that sensor placement has a significant impact on estimation accuracy. Validation with random forest (RF) models confirms the same trend as ANN, supporting the reliability of the findings.