Title:Angular velocity and linear acceleration measurement bias estimators for the rigid body system with global exponential convergence

Abstract:Rigid body systems usually consider measurements of the pose of the body using onboard cameras/LiDAR systems, that of linear acceleration using an accelerometer and of angular velocity using an IMU. However, the measurements of the linear acceleration and angular velocity are usually biased with an unknown constant or slowly varying bias. We propose a measurement bias estimator for such systems under assumption of boundedness of angular velocity. We also provide continuous estimates to the state of the system, i.e. the pose, linear velocity, and position of the body. These estimates are globally exponentially convergent to the state of the rigid body system. We propose two bias estimators designed with the estimate of the pose in the ambient Euclidean space of the Special Euclidean group and show global exponential convergence of the proposed observers to the state of the system. The first observer assumes knowledge of bounds of the angular velocity, while the second observer uses a Riccati observer to overcome this limitation. We show the convergence with an example of a rigid body rotation and translation system on the special Euclidean group. We show that the observer is able to estimate the bias using data collected from an Intel Realsense camera.