Robust Pose Estimation of Moving Objects Using Laser Camera Data for Autonomous Rendezvous & Docking

Different perception systems are available for the estimation of the pose (position and orientation) of moving objects. For space applications, an active vision system such as Laser Camera System (LCS) developed by Neptec Design Group (Ottawa, Canada) is preferable for its proven robustness in harsh lighting conditions of space. Based on LCS data, this paper presents results of integration of a Kalman filter (KF) and an Iterative Closest Point (ICP) algorithm in a closed-loop configuration. The initial guess for the ICP is provided by state estimate propagation of the Kalman filer. This way, the pose estimation of moving objects becomes more accurate and reliable in case when LCS does not deliver reliable data for a number of frames and the last known pose, used as an initial guess for the next one, is outside the ICP convergence range. In this case, the proposed algorithm automatically relies more on the dynamics model to estimate the pose, and vice versa. The Kalman filter, as a part of the integrated framework, is capable of not only estimating the target’s states, but also its inertial parameters. The convergence properties of this framework are demonstrated by experimental results from real-time scanning of a satellite model attached to a manipulator arm, which is driven by a simulator according to orbital and attitude dynamics. These results proved robust pose tracking of the satellite only if the Kalman filter and ICP are in the closed-loop configuration.