| Sumario: | The article provides information on the innovative tests developed by the Red Team of Carnegie Mellon University that were effective tools for measuring the driving skill of autonomous ground vehicles (AGV) in the 2005 Defense Advanced Research Projects Agency Grand Challenge in the U.S. The blind path-tracking, perception-tracking, and perception-planning tests evaluate how the unit changes in hardware and software affected the driving ability of the robots. The blind-tracking test is useful in measuring the ability of an AGV to blindly follow waypoints, while perception-tracking is a good tool for measuring the effects of perception on path-tracking and the perception-planning test is useful in measuring the capability of AGV to dynamically adapt to items impeding the preplanned path.
|
|---|