Investigations on robot drivers for vehicle exhaust emission measurements in comparison to the driving strategies of human drivers

摘要

New Federal and California Regulations present some new challenges for emission testing as low emission variability and different test cell environments through SFTP. One approach to achieve these challenges could be replacing the human driver by a robot driving system. To make a great step forward in the improvement of such robot systems, the German automotive technology research association (FAT) initialized an investigation program in cooperation with three different suppliers. The work was done and reported by the Department of Internal Combustion Engines at the Darmstadt University of Technology. This report summarizes the comparison of the driving style of the human driver to three automatic driving systems from those major manufacturers and some basic optimization work. The report also shows the results of investigations on six human test cell drivers with different unknown vehicles, focusing on the examination of the learning and driving strategies and the acting in specific situations as, for example, gear change, which are important topics for the driver models of automatic driving systems. Robot drivers are also of interest for AIGER for reducing test vehicle emission variability.