Ernst Dickmanns

From Wikipedia, the free encyclopedia

Ernst Dieter Dickmanns (born 1936), a former professor at Bundeswehr Universität München in Munich (since 1975), is the pioneer of Driverless cars. He studied aerospace and aeronautics at RWTH Aachen, and held positions at Princeton University, DFVLR Oberpfaffenhofen, and the NASA-Marshall Space Flight Center.

In the beginning of the 1980s his team equipped a Mercedes-Benz van with a camera and other sensors. The van was re-engineered such that it was possible to control steering wheel, throttle, and brakes through computer commands. Software was written that translated the sensory data into appropriate driving commands. For safety reasons, initial experiments in Bavaria took place on streets without traffic. Soon the Robot Benz managed to drive all by itself, at up to 97km/h, or roughly 60mph.

One of the greatest challenges in high-speed autonomous driving arose through the rapidly changing visual street scenes. Back then computers were much 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 active vision. Artificial saccadic movements allowed the system to focus its attention on the most relevant details of the visual input. Kalman filters were used to achieve robust autonomous driving even in presence of noise and uncertainty.

Partially encouraged by Dickmanns' success, in 1987 the EU's Eureka programme initiated the Prometheus project on autonomous vehicles, funded to the tune of nearly 800 million Euros. Most of the major car companies participated; so did Dickmanns and his team. Substantial progress was made in the following 7 years. In particular, Dickmanns' robot cars learned to drive in traffic under various conditions. An accompanying human driver with a "red button" made sure the robot vehicle could not get out of control and become a danger to the public. Transputers, a special breed of parallel computers, were used to deal with the (by 1990s standards) enormous computational demands.

A culmination point was achieved in 1995, when Dickmanns´ re-engineered autonomous S-Class Mercedes-Benz took a 1758km trip from Munich in Bavaria to Odense in Denmark and back. On highways the robot achieved speeds exceeding 175km/h (roughly 110mph; there is no general speed limit on the German Autobahn). Publications from Dickmann's research group, e.g. [1] indicate a mean autonomous distance between human interventions of 9km, with the longest autonomous stretch reaching 66km in Denmark, where a general speed limit of 110km/h is in place. This is particularly impressive considering that the system used a passive video-camera based sensor and did not model situations like road construction, lane-changes at over 140km/h, and other traffic with more than 40km/h relative speeds. In total, 95% autonomous driving (by distance) was achieved.

Unlike the early robot cars, the S-Class Robot drove in traffic. Therefore it had to recognize other moving objects and estimate their speeds. It even suggested and executed maneuvers to pass other cars. Unlike later robot cars, it used the approach to localize the robot laterally on the current road and follow it until instructed otherwise. Therefore, it did not localize itself in global coordinates and could drive without incorporating GPS and road maps as found in an automotive navigation system. The car's abilities left a big impression on many observers, and heavily influenced robot car research and funding decisions world-wide.