Passive navigation using egomotion estimates

The goal of this work is to propose a method to solve the problem of passiv e navigation with visual means. Passive navigation is the ability of an aut onomous agent to determine its motion with respect to the environment. The two main egomotion parameters allowing performing passive navigation are th e heading direction and the time to collision with the environment. A lot o f approaches have been proposed in literature in order to estimate the abov e parameters, most of which work well only if the motion is a predominant f orward translation and small amounts of noise are present in the input data . The method we propose is a two-stare approach: matching of features extract ed from 2D images of a sequence at different times and egomotion parameter computation. Both algorithms are based on optimization approaches minimizin g appropriate energy functions; The novelty of the proposed approach is to formulate the matching energy function in order to englobe invariant cues o f the scent. The matching stage recovers correspondences between sparse hig h interest feature points of two successive images useful to perform the se cond stage of egomotion parameter estimation. Experimental results obtained in real context show the robustness of the method. (C) 2000 Elsevier Scien ce B.V. All rights reserved.