|
Abstract : |
that autonomously navigating vehicles can bring to our future. Semiautomated traffic will dramatically increase throughput on existing highways, reducing our need for financially and environmentally expensive new roadways (as shown by the Nat?l Automated Highway Systems Consortium Demonstration ?97). The resulting reduction in commute time will mean a reduction in pollution, and former drivers will be free for more productive tasks. They will also be safer: autonomous vehicle systems or driver-assistance systems will help eliminate accidents caused by fatigue and human error. From the military perspective, sending driverless vehicles onto the battlefield will reduce the risk of human injury. In addition, because such vehicles won?t need armor, they will be lighter and faster. With these benefits in mind, the German and US departments of defense have sponsored the AutoNav program, 1 a joint effort to develop an autonomous vehicle for off-road navigation. This article describes the achievements of the program?s first year and the status of the vehicle prototype. Particularly, we focus on the integration of a top-down 4D perception and control architecture with bottom-up pyramid- and horopter-based imageprocessing techniques. Further, we describe, |
