After five vehicles completed the 2005 Grand Challenge, led by a team from Stanford University, DARPA determined that the problem of off-road autonomous navigation had been solved for military purposes.

The problem of on-road autonomous navigation within an urban setting, however, remains unsolved. Therefore, in May of 2006, DARPA announced the creation of The Urban Challenge. This new competition stresses the difficulties of making an autonomous vehicle drive within a complex urban network, including loss of GPS coverage, intersections, lane changing, merging, and parking - all while obeying proper traffic laws.

PAVE participated in the Urban Challenge with Prospect 12, a modified Ford Escape Hybrid fitted with several stereo and monocular cameras. Although PAVE was one of the 35 teams invited to participate in the National Qualification Event, Our vehicle did not compete in the final Urban Challenge race.

PAVE continues to use Prospect 12 as a test platform for research. Our approach to the Urban Challenge is summarized in the pages in this section. In general, our approach to problems in robotics can be briefly explained in terms of perception, cognition, actuation, and substrate.

Our robot must detect lanes and obstacles of various kinds (perception), fuse that data and select a new desired path (cognition), and drive along the new path (actuation). As the car follows its path its environment will change, presenting a new set of lanes and obstacles for perception. This processing loop depends on computers and power systems (substrate).