Odin: Team VictorTango's entry in the DARPA Urban Challenge

Abstract

Abstract The DARPA Urban Challenge required robotic vehicles to travel more than 90 km through an urban environment without human intervention and included situations such as stop intersections, traffic merges, parking, and roadblocks. Team VictorTango separated the problem into three parts: base vehicle, perception, and planning. A Ford Escape outfitted with a custom drive‐by‐wire system and computers formed the basis for Odin. Perception used laser scanners, global positioning system, and a priori knowledge to identify obstacles, cars, and roads. Planning relied on a hybrid deliberative/reactive architecture to analyze the situation, select the appropriate behavior, and plan a safe path. All vehicle modules communicated using the JAUS (Joint Architecture for Unmanned Systems) standard. The performance of these components in the Urban Challenge is discussed and successes noted. The result of VictorTango's work was successful completion of the Urban Challenge and a third‐place finish. © 2008 Wiley Periodicals, Inc.