High Precision Automotive Radar Target Simulation

Radar target simulators (RTSs) deceive a radar under test (RuT) by creating an artificial environment consisting of virtual radar targets. In this work, new techniques are presented that overcome the rasterization deficiency of current RTS systems and enable the generation of virtual targets at arbitrary high-precision positions. This allows for continuous movement of the targets and thus a more credible simulation environment.

During the evolvement of autonomous driving technology, obtaining reliable 3-D environmental information is an indispensable task in approaching safe driving. The operational behavior of automotive radars can be precisely evaluated in a virtual test environment by modeling its surrounding, specifically vulnerable road users (VRUs). Such a realistic model can be generated based on the radar cross section (RCS) and Doppler signatures of a VRU. Therefore, this work proposes a high-resolution RCS measurement technique to determine the relevant scattering points of different VRUs.

Accompanied with the development of the wireless communication technologies, the high data traffic is more necessary for civil and industrial applications than ever The concept of an intelligent reflective surface (IRS) has attracted considerable attention recently as a low-cost solution. As the main contribution, the dissertation creates new state-of-the-art and formulates a solid milestone for the IRS research field.

The simulation of the vehicle’s environmental sensors, the so-called sensor simulation, is crucial for testing and validating autonomous driving. Automobile manufacturers are increasingly focusing on a standardized architecture with a high level of abstraction. In order to simulate the sensors, such as radar sensors, most realistically on a point cloud level, data-based methods are used in many cases. In general, and specifically in case of radar sensors, there are still challenges to be faced. Therefore, four research questions are addressed: Is it possible to generate synthetic training data for data-based models? Which statistical approaches are suitable to simulate radar point clouds and how shall their learning capacities be evaluated? Is there a modeling approach to circumvent the disadvantages of statistical modeling? How to tackle the statistical nature of radar sensors during validation? Die Simulation der Umfeldsensoren des Fahrzeugs, die sogenannte Sensorsimulation, ist für Test und Absicherung des autonomen Fahrens entscheidend. Die Automobilhersteller setzen dabei zunehmend auf eine standardisierte Architektur mit hohem Abstraktionsgrad. Um die Sensoren, wie z.B. Radarsensoren, möglichst realitätsnah auf Punktwolkenebene zu simulieren, werden in vielen Fällen datenbasierte Methoden eingesetzt. Im Allgemeinen und speziell im Fall von Radarsensoren gilt es noch immer zahlreiche Herausforderungen zu meistern. Daher werden in dieser Arbeit vier Forschungsfragen behandelt: Können synthetische Trainingsdaten für datenbasierte Modelle generiert werden? Welche statistischen Ansätze sind geeignet, um Radar-Punktwolken zu simulieren und wie können die Ansätze bewertet werden? Gibt es einen Modellierungsansatz, um Nachteile der statistischen Modellierung zu umgehen? Wie kann die statistische Natur bei der Validierung berücksichtigt werden?

This book brings together papers presented at the 2022 International Conference on Communications, Signal Processing, and Systems, online, July 23-24, 2022, which provides a venue to disseminate the latest developments and to discuss the interactions and links between these multidisciplinary fields. Spanning topics ranging from communications, signal processing and systems, this book is aimed at undergraduate and graduate students in Electrical Engineering, Computer Science and Mathematics, researchers and engineers from academia and industry as well as government employees (such as NSF, DOD and DOE).

This book constitutes the refereed proceedings of the 11th International Conference on Computer Vision Systems, ICVS 2017, held in Shenzhen, China, in July 2017. The 61 papers presented were carefully reviewed and selected from 92 submissions. The papers are organized in topical sections on visual control, visual navigation, visual inspection, image processing, human robot interaction, stereo system, image retrieval, visual detection, visual recognition, system design, and 3D vision / fusion.

This is the first book to bring together the increasingly complex radar automotive technologies and tools being explored and utilized in the development of fully autonomous vehicles – technologies and tools now understood to be an essential need for the field to fully mature. The book presents state-of-the-art knowledge as shared by the best and brightest experts working in the automotive radar industry today -- leaders who have “been there and done that.” Each chapter is written as a standalone "master class" with the authors, seeing the topic through their eyes and experiences. Where beneficial, the chapters reference one another but can otherwise be read in any order desired, making the book an excellent go-to reference for a particular topic or review you need to understand. You’ll get a big-picture tour of the key radar needs for fully autonomous vehicles, and how achieving these needs is complicated by the automotive environment’s dense scenes, number of possible targets of interest, and mix of very large and very small returns. You’ll then be shown the challenges from – and mitigations to – radio frequency interference (RFI), an ever-increasing challenge as the number of vehicles with radars – and radars per vehicle grow. The book also dives into the impacts of weather on radar performance, providing you with insights gained from extensive real-world testing. You are then taken through the integration and systems considerations, especially regarding safety, computing needs, and testing. Each of these areas is influenced heavily by the needs of fully autonomous vehicles and are open areas of research and development. With this authoritative volume you will understand: * How to engage with radar designers (from a system integrator / OEM standpoint); * How to structure and set requirements for automotive radars; * How to address system safety needs for radars in fully autonomous vehicles; * How to assess weather impact on the radar and its ability to support autonomy; * How to include weather effects into specifications for radars. This is an essential reference for engineers currently in the autonomous vehicle arena and/or working in automotive radar development, as well as engineers and leaders in adjacent radar fields needing to stay abreast of the rapid developments in this exciting and dynamic field of research and development.

Advances in Electromagnetics Empowered by Artificial Intelligence and Deep Learning Authoritative reference on the state of the art in the field with additional coverage of important foundational concepts Advances in Electromagnetics Empowered by Artificial Intelligence and Deep Learning presents cutting-edge research advances in the rapidly growing areas in optical and RF electromagnetic device modeling, simulation, and inverse-design. The text provides a comprehensive treatment of the field on subjects ranging from fundamental theoretical principles and new technological developments to state-of-the-art device design, as well as examples encompassing a wide range of related sub-areas. The content of the book covers all-dielectric and metallodielectric optical metasurface deep learning-accelerated inverse-design, deep neural networks for inverse scattering, applications of deep learning for advanced antenna design, and other related topics. To aid in reader comprehension, each chapter contains 10-15 illustrations, including prototype photos, line graphs, and electric field plots. Contributed to by leading research groups in the field, sample topics covered in Advances in Electromagnetics Empowered by Artificial Intelligence and Deep Learning include: Optical and photonic design, including generative machine learning for photonic design and inverse design of electromagnetic systems RF and antenna design, including artificial neural networks for parametric electromagnetic modeling and optimization and analysis of uniform and non-uniform antenna arrays Inverse scattering, target classification, and other applications, including deep learning for high contrast inverse scattering of electrically large structures Advances in Electromagnetics Empowered by Artificial Intelligence and Deep Learning is a must-have resource on the topic for university faculty, graduate students, and engineers within the fields of electromagnetics, wireless communications, antenna/RF design, and photonics, as well as researchers at large defense contractors and government laboratories.