A machine is anything that helps us simplify work. By that definition, a car is a machine that helps us simplify the work of locomotion. However, we do need to do some work when using a car – we need to drive it. Wouldn’t life be easier if we could just get into the car, say where we want to go and doze off until the destination is reached? Fantastical as it may seem, it may be possible in near future if the innovations showcases in the DARPA Urban Challenge in 2007 are developed to their full potentials.
If you are a science-fiction fan, you must remember the taxi with the robot driver in “Total Recall” that Arnold Schwarzenegger dismantles violently when it refuses his order to “just drive” and insists on knowing the name of his destination. Although you may not see such autonomous cars on the roads anytime soon, the participants at the DARPA Urban Challenge give us cause for hope.
While the first two required driverless cars to negotiate roads, turns and tunnels in the desert without human intervention, the last one required the cars to follow traffic laws in a simulated urban environment. Therefore, while the first two challenges were more physically demanding but had little interaction between the vehicles, the Urban Challenge required the individual participants to make “intelligent” decisions in real time based on the actions of other vehicles.
Although the competitions were open to teams across the world, each team had to have at least one American citizen as a member. The reason behind these competitions was the ultimate goal of making,one-third of US ground military forces autonomous by 2015.
The third competition was the toughest of all and required “teams to build an autonomous vehicle capable of driving in traffic, performing complex maneuvers such as merging, passing, parking, and negotiating intersections.” Development of such technology has potentially immense benefits not only for military purposes, but civilian uses as well. Besides the obvious advantage of ease of travel, this technology, when revised and popularized, can ease traffic congestion and prevent accidents, saving thousands of lives and billions of dollars every year in addition to reducing pollution and unproductive time.
In a nutshell, the participants were required to comply with the following rules:
Vehicle must be stock or have a documented safety record.
Vehicle must obey the California state driving laws.
Vehicle must be entirely autonomous, using only the information it detects with its sensors and public signals such as GPS.
DARPA will provide the route network 24 hours before the race starts.
Vehicles must traverse the route by driving between specified checkpoints.
DARPA will provide a file detailing the order the checkpoints must be driven to 5 minutes before the race start.
Vehicles may “stop and stare” for 10 seconds maximum.
Vehicles must operate in rain and fog, with GPS blocked.
Vehicles must avoid collision with vehicles and other objects such as carts, bicycles, traffic barrels, and objects in the environment such as utility poles.
Vehicles must be able to operate in parking areas and perform U-turns as required by the situation.
Of the 89 teams that applied, 53 were notified they had qualified for DARPA site visits based on their technical papers and videos. After actual testing of the vehicles, the list was whittled down to 35 who were invited to the National Qualification Event (NQE), a rigorous eight-day vehicle testing period. The NQE was held across three different test areas in the remodeled George Air Force Base (currently used as Southern California Logistics Airport), in Victorville, California. This was also the venue for the finals three months later to which 11 of the 35 at the NQE qualified.
These 11 finalists included a lot of collaborations between universities and their corporate partners – Carnegie Mellon with GM, Stanford with Volkswagen, Virginia Tech with TORC Technologies, etc. The final event consisted of three missions totaling around 55 miles, to be completed in six hours. Besides the competitors themselves, 30 Ford Tauruses driven by professional drivers were introduced into the course to increase complexity of traffic maneuvers.
While time to finish was a major factor in allocation of points, participants were penalized for violations, dangerous behavior, delays, etc. Thus, although the team from Stanford finished earliest with their vehicle “Junior,” a Volkswagen Passat, the team from Tartan Racing, a collaboration between Carnegie Mellon and GM, with their Chevrolet Tahoe “Boss,” were declared the winners after tabulating points. Of the 11 finalists, six finished the course within 6 hours, testimony to the tremendous improvements since the first Grand Challenge 3 years before where not a single vehicle finished the course.
In conclusion, all 11 finalists received DARPA funding for additional research and it is hoped that the lessons from these competitions have brought us nearer to a future of driverless cars.