Position, Velocity, Attitude and Accelerometer-Bias Estimation from IMU and Bearing Measurements

This paper considers the problem of estimating the position, attitude and velocity of a rigid-body in a 3D space by fusing bearing measurements provided by a monocular camera with gyroscopic and accelerometer measurements provided by an Inertial Measurement Unit (IMU). The proposed deterministic observer is accompanied with an observability analysis, that points out the minimum number of image points (bearings) along with their configuration in the inertial frame, under which local exponential stability is guaranteed. The performance of the observer is demonstrated by performing experiments on a test-bed inertial-Visual sensor.