MUNICH, Germany — The large red SUV bends around the corner and approaches the intersection. It stops, waits until another car has passed and then the driver accelerates to cross the intersection.

The driver? No driver can be seen on the seat behind the steering wheel; the car is empty. Instead, a vast array of cameras, antennas and odd-looking sensors is mounted on the car’s roof and its bumpers. The back seat is occupied by a stack of computers. Carefully accelerating, the car disappears around the corner.

This is the likely scene during the Defense Advanced Research Projects Agency (DARPA) Urban Challenge that takes place next month at the former George Air Force Base in Victorville, California. The Urban Challenge final event is scheduled for November 3rd. At the moment 36 semi-finalists are in the running, a number that is scheduled to be whittled down to 20 teams during a qualification stage taking place during the last week of October.

The Urban Challenge, a research project of the US Department of Defense, is targeted at the design of autonomously driving vehicles. In order to make the driver dispensable, the vehicles are equipped with instruments that basically replicate a human’s senses and brainpower.

This is far from being a trivial task. In contrast to earlier DARPA contests on autonomous driving, it will not be sufficient for the vehicles simply to find their way to a final destination. This time, the contest rules demand that the vehicles navigate through an urban environment over a total distance of 60 miles. This poses challenges such as merging into moving traffic, avoiding moving obstacles, finding the way through roundabouts and, finally, parking in a gap - all according to California traffic laws.

More sophisticated aspects of everyday driving such as recognition of traffic signs, traffic lights or pedestrians are not on the requirements list – for the time being.

While the rules of the contest demand that at least the leader of each team has to be a citizen and resident of the U.S., many teams from Europe and Japan are participating. For instance, Volkswagen is contributing staff and technology to the Stanford Racing Team which won the previous DARPA Grand Challenge.

Team AnnieWay is a spin-off of the Collaborative Research Center on Cognitive Automobiles, formed by the University of Karlsruhe, the Technical University of Munich, the University of the German armed forces and the Fraunhofer Gesellschaft. Team Berlin is a joint project of the Freie Universität Berlin with Fraunhofer Institute for Intelligent Analysis and Information Systems (IAIS) and the Rice University in Houston, Texas. The Team CarOLO includes students, academics and researchers from different institutes of the University of Braunschweig.

To succeed in navigating through the Victorville ‘urban jungle’, the vehicles will rely on a broad range of different sensors. Besides GPS for the navigation, the cars are equipped with optical cameras, radar, and a variety of light detection and ranging (lidar) systems and scanners that jointly simulate a human driver’s eyesight. The huge amount of input data is then fed into computers that generate the signals that effectively steer the cars.

In practice, lidar is the prevalent sensor type. Of the 36 teams, three-quarter use the infrared scanners of German optical systems vendor Sick AG (Waldkirch, Germany). But, of course, IR scanners alone are not enough. “A lidar system alone cannot distinguish between a car, a Coca Cola can and a biker,” quips Professor Ferdinand Dudenhöffer, professor for automotive technology at the Gelsenkirchen University of Applied Sciences.

Most teams agree with this assessment of a lidar’s capability, and have rigged their vehicles accordingly. “We use two different types of laser systems which, based on signal propagation time, compute a 3D image of the surroundings. But laser alone does not suffice,” notes Bernhard Rumpe, Professor at the Braunschweig University and one of the managers of Team CarOLO.

The group thus complements the lidar signals with radar systems. In addition, the CarOLO vehicle uses a camera for the lane recognition. “The laser system is not smart enough,” he says. “The algorithms make the difference.” The software developed at the Braunschweig university very accurately discriminates between size and motion vectors of objects.

Despite all this, the CarOLO solution is not yet perfect, Rumpe admits. “It still generates too many ‘false positive’ identifications, which have to be eliminated through plausibility checks.” The specific strength of his team is the multi-disciplinary approach, Rumpe asserts, referring to the fact that the assembled group includes experts from the university’s mechanical engineering department as well as from two software institutes associated with the university.

In this regard, the CarOLO team resembles most of the other competing teams with a strong academic and research background. However, the LUX team has followed a completely different approach. A joint project by Hamburg-based startup Ibeo Automobile Sensor GmbH and its parent company Sick AG, the LUX team relies solely on infrared sensors, testing the prototype of a new laser scanner generation that uses four beams instead of one.

Equipped with two scanners integrated in the front bumpers and one in the rear bumper, the vehicle stands out from the others in that no futuristic sensors are visible; the vehicle appears completely as if a normal volume car. Even the back seat – where all other contestants’ vehicles have installed an array of different dedicated computers – is empty in the LUX car.

“No cameras, no radar, no ultrasound sensors,” explains LUX marketing manager Tanja Müller. “We have managed to integrate the entire pre-processing into the sensors, reducing the amount of input data for the main computer. Thus, we get by with two small Pentium M-equipped Linux PCs in the trunk (or boot) for the generation of the steering and speed control,” she explains. “The sensor system, derived from a brake assistant, is mature enough to detect motion vectors of obstacles and even the nature of the obstacle – if it has legs, it is human,” she says.

Once the robot vehicles have done their “runoff” at the George Air ForceBase, the organizers as well as the participating teams will start to evaluate the results. They will likely come to different conclusions, because their motivation is different.

While the DARPA sees the Urban Challenge as developing the technologies and techniques that will eventually enable vehicles to operate autonomously in high-risk areas, the automotive industry is less interested in a development leading towards autonomous vehicles. “I do not believe that we will see autonomous cars in public traffic for the next 20 years,” says Professor Dudenhöffer. "But the cognition gained during the contest could be used for implementing features that increase traffic safety – for instance it could do away with the ‘blind area’ for trucks. And, these technologies could provide safety back-ups in an ageing society."

Robot eyes looking at you – with optical, infrared and radar sensors. Source: CarOLO Team

Automotive OEM Volkswagen, a major sponsor for several teams, has a similar point of view. “We want to find out which possibilities exist today. Our aim is not the vehicle that drives without a driver, but we want to develop more intelligent driver assistance systems – for instance, adaptive cruise control systems with ‘follow to stop’ function,” Volkswagen Technology spokesman Harthmuth Hoffmann explains. “Our target is refining driver assistance systems. And we believe this is a trend throughout the entire automotive industry.”