Valois et al. 1997 : Artificial vision has proven to be an important enabling technology for space robotics as evidenced by the success of the Canadian Space Vision System (SVS). Based on cooperative targets and photometric analysis, the system is able to track the position and orientation of targets such as sattelites in real-time. With advances in sensors and microelectronics, particularly in the area of high performance computing, many of the constraints present in the original system design have been relaxed to a significant degree. One possibility of particular interest is the elimination of cooperative targets, relying instead on the intrinsic structure of objects being tracked. The goal of this project is to investigate how techniques from dynamic scene analysis, e.g. optical flow, can be used to estimate the pose of a moving object passing in front of a stationary observer. Our related work in model-based tracking from optical flow provides a starting point. Given prior knowledge about the shape of a moving object, it becomes possible to remove some of the ambiguity inherent in the structure from motion problem. Other constraints, possibly derived from dynamic control of camera parameters (i.e. active vision), will be investigated to determine how this ambiguity can be fully resolved, leading to reliable pose estimates. [Baron T (1997)]