Ambulatory Hip Angle Estimation Using Multiple Model Hybrid Dynamic Bayesian Networks
Zhang Zhi-qiang①② Huang Zhi-pei① Wu Jian-kang①
①(Sensor Network and Application Research Center (SNARC), Institute of Automation and Graduate University, Chinese Academy of Sciences, Beijing 100190, China) ②(Department of Computing, Imperial College, London UK SW7 2AZ)
Abstract：Hip angle is a major parameter in gait analysis while gait analysis plays important role in healthcare, animation and other applications. Accurate estimation of hip angle using wearable inertial sensors in ambulatory environment remains a challenge, this is mainly because (1) the non-linear nature of thigh movement has not been well addressed, and (2) the variation of micro-inertial sensor measurement noise has not been studied yet. We propose to use Hybrid Dynamic Bayesian Network (HDBN) and multiple motion models and multiple noise models to model the non-linear hip angle dynamics and variation of measurement noise levels. Gaussian Particle Filter (GPF) is employed to estimate the hip angle during gait cycles from the measurements of accelerometers that are attached to the thighs. The experimental results show that the proposed method can achieve significant accuracy improvement over the previous work on the ambulatory hip angle estimation.
2011, Vol. 33 
