Foreword for Journal of Field Robotics—Special Issue on the DARPA grand challenge

Some call it the race of the century. The Defense Advanced Research Projects Agency DARPA has a rich history of pursuing innovative technical ideas that lead to capabilities never before dreamed possible. DARPA’s past projects resulted in the ARPANET which led to the Internet , stealth, and the Predator and Global Hawk unmanned aerial vehicles. In 2003, Dr. Tony Tether, DARPA Director, believed the time was right to push autonomous ground vehicle technology. The thinking was that if such vehicles were possible, they could be used in dangerous military missions and save the lives of young American men and women in uniform. At that time, many experts believed fully autonomous ground vehicles that could travel great distances at speeds important to military operations were not possible within any near-term timeline. Thus, Dr. Tether created the DARPA Grand Challenge as an experiment to accelerate autonomous ground vehicle technology. The expectation was that only a few people would respond, but the Grand Challenge awakened a new surge of creativity among thousands from the United States and other countries. The first Grand Challenge was held in March 2004 and, from an initial field of 106 applicants, 15 teams competed in the final event. The best vehicle managed to travel 5% of the route before failing. It was clear then that the challenge was indeed “grand.” At the 2004 closing ceremony, DARPA announced another competition and doubled the prize to $2 million. October 2005 was chosen for the second competition—a mere 18 months after the first event. America’s innovators again rolled up their sleeves and got to work. Eighteen months is not a long time to create technology that may change the world. DARPA received 197 applications to compete in the 2005 event, and each of the 23 teams that made it to the final event had a vehicle that proved, through a series of difficult qualifying trials, to be better than the best vehicle in the 2004 race. The 2005 course was tough: 132 miles of difficult desert roads across Nevada that contained a mixture of featureless terrain, dust, global positioning system drop-outs, sharp turns, narrow openings, bridges, railroad overpasses, long tunnels, obstacles, and a narrow winding mountain road with a 200-foot drop-off. The actual route was kept secret until 2 hours before the start. Remarkably, five teams finished the course, four of them under the 10-hour limit and within 37 minutes of the winning time of Stanford’s “Stanley” at 6 hours and 53 minutes and an average speed of 19.2 miles per hour. In another first for autonomous vehicle operations, “TerraMax” finished the course on the second day after remaining parked overnight in autonomous mode. The vehicles that did not finish the course suffered mechanical, system, or software problems. In the end, 22 of the 23 teams traveled farther than the best vehicle did in 2004. By all accounts, Grand Challenge 2005 was a spectacular success. The primary reason for the success of the Grand Challenge lies with the teams—the students, engineers, scientists, and backyard mechanics—all inventors who brought fresh ideas to solve a very difficult technical problem. They were individuals, but learned to work together in teams. Unlike other prize events, the DARPA Grand Challenge set a specific date for the competition, and all the teams spent countless long hours overcoming setbacks to be there for the final event. The teams accelerated autonomous ground ve• • • • • • • • • • • • • • • • •