Metric Velocity, Landmark Distance and Attitude Estimation utilizing Monocular Camera Images and Gravity biased IMU Data

In this paper we present a novel framework for metric velocity, landmark distance and attitude estimation for partially accelerated trajectories in the presence of gravity. The proposed approach relies solely on monocular camera images and measurements from an inertial measurement unit with 9 degrees of freedom. Furthermore it does not require any prior information or assumptions regarding the desired estimates or the static environment. This is achieved by utilizing a novel method for attitude estimation in the filter framework and by applying a novel method for filter initialization. The capabilities and the performance of the resulting navigation filter are demonstrated through simulations.