Estimating the Orientation of a Game Controller Moving in the Vertical Plane Using Inertial Sensors
This paper presents a novel method of estimating the orientation of a rigid body in the vertical plane from point-acceleration measurements, by discerning its gravitational and inertial components. In this method, a simple stochastic model of the human-hand motions is used in order to distinguish between the two types of acceleration. Two mathematical models of the rigid-body motion are formulated as distinct state-space systems, each corresponding to a proposed method. In both two cases, the output is a nonlinear function of the state, which calls for the application of the extended Kalman filter (EKF). The proposed filter is shown to work efficiently through two simulated trajectories, which are representative of human-hand motions. A comparison of the orientation estimates obtained from the proposed method shows that the filter offers more accuracy than a tilt sensor under high accelerations, and avoids the drift obtained by the time-integration of gyroscope measurements.