Evaluation of Iterative Missile Pose Estimation Algorithm using Point Pattern Matching for Terminal Guidance and Studies on Convergence Criteria

Missile pose estimation in terminal guidance refers to the technique of estimating missile position and attitude by matching seeker images with pre stored reference images during the terminal phase of missile flight. This paper presents a method for estimating the position and orientation of the missile with respect to a known object using point correspondences between seeker and reference images. Seeker view is perspective in nature and hence there exists a nonlinear relationship between the points in the seeker image and the ground control points in the reference image. Hence by solving a set of linear equations, it is not possible to estimate missile pose directly from point correspondences. Assuming a scaled orthographic projection which is linear in nature, one can estimate initial approximate missile pose. In the present work we follow an iterative scheme for estimating missile pose starting from initial orthographic assumption. We describe a convergence criteria and establish a scheme that allows missile position x, y, z and heading ψ, θ, φ estimation in about 3-5 iterations. At convergence the result is compatible with perspective camera model. In this paper, we also describe in detail an iterative pose computation for coplanar object points. We present the results showing the sensitivity of the algorithm with varying trajectory parameters like altitude, range and position\ heading errors.