<p>Increasing interest in space debris removal and on-orbit servicing have motivated research in autonomous relative navigation algorithms using Light Detection And Ranging (LIDAR) sensors. While relative navigation schemes are well-established for cooperative target spacecraft, the more practical but challenging problem of a noncooperative and unidentified target is considered in this paper. That is, the target does not furnish any navigational aids to the chaser, and no pre-existing model of the target geometry is accessible to the navigation system. To reconstruct the target and track the relative states, an Extended Target Tracking (ETT) technique is implemented in which a weighted sum of spherical harmonics basis functions models the target surface. As LIDAR point clouds are collected, the Multiplicative Extended Kalman Filter estimates the weights simultaneously with the 6-degree-of-freedom relative states. Results of simulated rendezvous with a CubeSat show successful target shape estimation and tracking of relative kinematics, position, and attitude.</p>