Share Email Print

Proceedings Paper

Obstacle Avoidance On The Autonomous Land Vehicle Using The Warp Systolic Array Machine
Author(s): R. Terry Dunlay
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

This paper describes range image based obstacle avoidance methods developed to run on the Warp Machine, a new linear systolic array processor capable of delivering up to 100 million floating point operations per second. The Warp processing described here has been tested on the DARPA Autonomous Land Vehicle (ALV) to detect and locate obstacles present in an outdoor road environment at speeds up to 10 km/hour. The overall perception system consists of two subsystems; a color image processing subsystem used to find the road area and the Warp based range image processing subsystem. We briefly describe the architecture of the Warp Machine and methods of exploiting the parallelism of the Warp system. Next we describe the details of the obstacle avoidance perception algorithm and its mapping onto the Warp Machine. Results of applying the system in an experimental test run are presented.

Paper Details

Date Published: 1 January 1987
PDF: 10 pages
Proc. SPIE 0852, Mobile Robots II, (1 January 1987); doi: 10.1117/12.968235
Show Author Affiliations
R. Terry Dunlay, Martin Marietta Astonautics Group (United States)

Published in SPIE Proceedings Vol. 0852:
Mobile Robots II
Wendell H. Chun; William J. Wolfe, Editor(s)

© SPIE. Terms of Use
Back to Top