VaMP

From Wikipedia, the free encyclopedia

The VaMP robot car was possibly the most influential vehicle in the history of driverless cars, together with its twin vehicle VITA-2. They were the first truly autonomous cars, able to drive in heavy traffic for long distances without human intervention, using computer vision to recognize rapidly moving obstacles such as other cars, and automatically avoiding and passing them.

The VaMP was constructed by the team of Ernst Dickmanns at Bundeswehr Universität München and Mercedes-Benz in the 1990s as part of the 800 million ECU EUREKA Prometheus Project on autonomous vehicles (1987-1995). It was a 500 SEL Mercedes re-engineered such that it was possible to control steering wheel, throttle, and brakes through computer commands based on real-time evaluation of image sequences. Software was written that translated the sensory data into appropriate driving commands. Back then, computers were over 100 times slower than they are today; therefore, sophisticated computer vision strategies were necessary to react in real time. The team of Dickmanns solved the problem through an innovative approach to dynamic vision. Attention control including artificial saccadic movements of the platform carrying the cameras allowed the system to focus its attention on the most relevant details of the visual input. Four cameras with two different focal lengths for each hemisphere were used in parallel for this purpose. Kalman filters were extended to handle perspective imaging and to achieve robust autonomous driving even in presence of noise and uncertainty. Several dozen Transputers, a special breed of parallel computers, were used to deal with the (by 1990s standards) enormous computational demands.

In 1994 the VaMP and its twin Vita-2 were stars of the final international presentation of the PROMETHEUS project in October 1994 on Autoroute 1 near the airport Charles-de-Gaulle in Paris. With guests onboard, the twins drove more than 1000 km in heavy traffic on the three-lane highway at speeds up to 130 km/h. They demonstrated lane changes left and right, autonomously passing other cars.

One year later the autonomous Mercedes-Benz drove more than 1,000 miles (2,000 km) from Munich to Copenhagen and back, in traffic, at up to 175 km/h, again planning and executing maneuvers to pass other cars. Only in a few critical situations (such as unmodeled construction areas) a safety driver took over. Again, active computer vision was used to deal with rapidly changing street scenes. The robot achieved speeds exceeding 175 km/h on the German Autobahn, with a mean time between human interventions of 9 km. Despite being a research system without emphasis on long distance reliability, it drove up to 158 km without any human intervention.

Most current robot cars as well as commercial vehicles with driver assistance use GPS such that they "know" precisely where they are. Remarkably, the VaMP used vision only, which is much harder.

VaMP and VITA-2 pioneered many hardware and software concepts that are essential for autonomous robots. They left a big impression on many observers and heavily influenced robot car research and funding decisions world-wide.

[edit] Related Topics

Languages