Incremental displacement estimation (IDE) for structural health monitoring robot

To measure 6-DOF displacement, a visually servoed paired structured light system (ViSP) has been introduced which is composed of two sides facing with each other, each with one or two lasers, a camera, a screen, and a 2-DOF manipulator. In the system, the lasers on both two sides project their parallel beams to the screen on the opposite side and 6-DOF relative displacement between two sides can be calculated using the positions of the three projected laser beams and the rotation angles of the manipulators. In the estimation process, extended Kalman filter (EKF) or Newton-Raphson method have been used. However, these aforementioned iterative methods have a drawback that the computation time is relatively long due to the multiple iterations. Therefore, in this paper, incremental displacement estimation (IDE) algorithm which updates the previously estimated displacement using the difference of the previous and the current observed data is introduced. To validate the performance of the newly proposed algorithm, various experimental tests have been performed. The results show that the ViSP with the IDE algorithm estimates the 6-DOF relative displacement between two sides with the same level of accuracy of EKF with multiple iterations in significantly reduced time.