Tightly Coupled Visual-Inertial Navigation System Using Optical Flow

This paper presents a state estimation framework that allows estimating the attitude, full metric speed and the orthogonal metric distance of an IMU-camera system with respect to a plane. The filter relies on optical flow- as well as gyroscope and accelerometer measurements. The underlying assumption is that the observed visual feature lies on a static plane. The orientation of the observed plane is not required to be known a-priori and is also estimated at run-time. The estimation framework fuses visual and inertial measurements in an Extended Kalman Filter (EKF). Experiments using a hand-held visual-inertial sensor successfully demonstrate the performance of the filter. It is shown that the state estimate is converging correctly, even in presence of substantial initial state errors. The minimal sensor suite, which is both light-weight and low-cost, renders the framework an appealing choice for the use as a navigation system on a wide range of robotic platforms, such as ground- or flying robots.