Showing papers by "Larry Matthies published in 1986"

Error Modelling in Stereo Navigation.

Abstract: In stereo navigation, a mobile robot estimates its position by tracking landmarks with on-board cameras. Previous systems for stereo navigation have suffered from poor accuracy, in part because they relied on scalar models of measurement error in triangulation. This paper shows that using 3-D gaussian distributions to model triangulation error leads to much better performance. The paper describes how to compute the error model from image correspondences, estimate robot motion between frames, and update the global positions of the robot and the landmarks over time. Simulations show that compared to scalar error models the 3-D gaussian reduces the variance in robot position estimates and better distinguishes rotational from translational motion. A short indoor run with real images supported these conclusions and computed the final robot position to within 2% of distance and one degree of orientation. These results illustrate the importance of error modelling in stereo vision for this and other applications. This research was sponsored by the Office of Naval Research under contract N00014-81 -K-0503 and by the Defense Advanced Research Projects Agency under contracts DACA 76-35-C-0003 and F33615-84-K-1520. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of ONR, DARPA, or of the U.S. Government