Data Driven Radar Detection Models: A Comparison of Artificial Neural Networks and Non Parametric Density Estimators on Synthetically Generated Radar Data
Thomas Eder, Rami Hachicha, Houssem Sellami, Carlo van Driesten, Erwin M. Biebl · mediaTUM – the media and publications repository of the Technical University Munich (Technical University Munich) · 2019
A rapid and agile development in the field of autonomous driving requires in particular the use of simulation. In order to ensure functional safety even in consequence of sensory defects, measurement deviations, or an incorrect environment model it is necessary to implement sensor models. Various data driven sensor modeling approaches, which are of particular interest for car manufacturers, have been introduced. In this paper we compare the learning capabilities of two different modeling approaches: Deep generative networks and non parametric density estimators. For comparative purposes and to give a detailed and realistic insight of existing strengths and weaknesses of each approach we use an, beyond the current state of the art, algorithm to generate a synthetic dataset.