The Defense Advanced Research Projects Agency (DARPA) challenged autonomous ground vehicle developers in the "2005 DARPA Grand Challenge" to build a vehicle that could complete a 132 mile course through the American desert southwest. MITRE, a not-for-profit systems engineering company, responded to this challenge by creating the MITRE Meteor in just 11 months. This rapid development relied on software employment transparency to get the maximum utility out of each line of code. Judicious design of the software framework allowed the same body of code to animate the robot in the field, support laboratory experimentation, and analyze recorded field testing data. This paper describes how software employment transparency was achieved and how it increased development efficiency.
Unfortunately, ACM prohibits us from displaying non-influential references for this paper.
To see the full reference list, please visit http://dl.acm.org/citation.cfm?id=1218349.