For a better experience on dSPACE.com, enable JavaScript in your browser. Thank you!

Presentation Abstracts

Keynote Address: How Will Autonomous Driving Transform dSPACE?

André Rolfsmeier, Lead Product Manager, Advanced Applications and Technologies, dSPACE GmbH

To master the enormous challenges with autonomous driving, the automotive industry is facing a transformation from engineering to high-tech IT companies. Similarly, dSPACE is called upon to deal with new and rapidly evolving topics. This presentation addresses the main challenges with autonomous driving and outlines the dSPACE strategy in this regard. Issues like sensor-realistic simulation, new testing methodologies, simulation in the cloud, and the role of artificial intelligence at dSPACE will be addressed.

Keynote Address: dSPACE Automotive Radar Test Systems - Over-the-Air Simulation of Echoes for Automotive Radar Sensors

Michael Rozmann, Executive Director and Owner of miro-sys
Dr. Andreas Himmler, Sr. Product Manager, Hardware-in-the-Loop Testing System, dSPACE GmbH

Radar sensors play an important role when it comes to integrating intelligent solutions for assisted and autonomous driving in vehicles. dSPACE Automotive Radar Test Systems (DARTS) can efficiently ensure the correct functioning of sensors and developed application software in different phases of the development, production and release process for sensors and vehicles, as well as for aftermarket use.

Scenario–Based Testing (SCbT) – Real Benefits of Scalability, Flexibility and Efficiency in an End-to-End Toolchain

Jace Allen, Business Development Manager, dSPACE Inc.

Validating highly complex autonomous systems is a huge challenge for any company. Given the myriad of traffic situations and scenarios for testing, infrastructure and resources required for testing, it can be quite expensive and time consuming to handle validating these systems with solid regression capabilities.  In order to develop and test ADAS/HAD systems, it is critical to utilize both virtual validation with virtual ECUs (V-ECUs) and hardware-in the-loop (HIL) to get the full bandwidth required for verification and validation (V&V).  It is also critical to be able to manage the daunting amount of tests and test assets required for the end-to-end processes. 

This session will present a complete toolchain to handle your AV testing capabilities and quality assurance using an automated simulation and test environment that interfaces to or simulates ADAS/AV sensor technology and other key technologies such connectivity, vehicle buses, and AV system frameworks. dSPACE scenario-based testing (SCbT) technology provides a strong foundation to build a continuous integration process for MIL, SIL, and HIL testing that can take advantage of the scalability of cluster and cloud testing.  The key to performing such virtual and real-time testing is the ability to stimulate the real or virtual ECUs and sensor interfaces through closed-loop simulation of the vehicle, its environment, traffic, surroundings, etc., along with playback of captured sensor data and its synchronization with key vehicle connectivity systems.  This presentation will also address effective techniques for the optimization of test processes and resources that fully enable autonomous test development and implementation.

TRACK 1 – Easing the Development Challenge

dSPACE Prototyping Systems – The Fastest Way to Autonomous Driving

Christian Woerdehoff, Product Manager – Rapid Prototyping Systems, dSPACE GmbH

Bringing highly automated vehicles on the road is the vision of many manufactures in the automotive domain. To make this vision come true, challenging requirements such as 360° redundant surround view with numerous heterogeneous sensors, high-precision positioning and real-time control have to be addressed. To address these challenges, dSPACE offers unique rapid prototyping solutions based on high-performance platforms and tailored software environments.    

RTMaps for Recording, Tagging, Iterating, and Accelerating Perception Development

Jacob Perrin, Application Engineer, dSPACE Inc.

RTMaps from Intempora is the most optimized, effective tool for the development of perception software applications. In this presentation, we will discuss new developments of the tool, including the RTag extension – a tablet interface for adding metadata during sensor data collections. Also included as part of this presentation will be a variety of live demos.

 

Robust Analytics and Benchmarking of Automated Driving Features

Modar Horani, Managing Principal, P3 North America

With the increased complexity of Automated Driving Systems (ADS), the need is growing to derive standard measures of quality and confidence in system performance. The P3 team is actively working to develop robust solutions and platforms supporting the future of ADS. This presentation will discuss the details of a modular platform that allows for analyzing ADS features on various vehicle configurations.

TRACK 2 – Modeling for Simulation

Keeping It Real - Simulating the Real World for Autonomous Driving System Validation

Thorsten Oppermann, Account Manager, dSPACE Inc.

The definition of meaningful simulation scenarios includes road descriptions, environment, vehicle and traffic parameterizations, as well as their movements. These are the key factors for successful virtual validation of highly automated driving functions. The dSPACE simulation tool chain ASM (Automotive Simulation Models) offers a variety of possibilities for road and scenario definitions based on requirements, real world data, and standards, like OpenDrive, OpenScenario and OSI. This presentation will also introduce the newly integrated functionality of ModelDesk Testing, making it an easy to use validation tool chain.

Sensor Realistic Simulation - Bringing the Environment for Testing

Patrik Moravek, Product Engineer, HIL Testing Systems, dSPACE GmbH

A detailed and comprehensive simulation of the real world is the challenge for the successful validation of autonomous driving system software. In addition to simulation models of the automated vehicle, traffic participants, etc., a wide range of sensor models are also utilized. The types of sensor models extend from technology-independent variants, which generate object lists directly from the information provided by the environmental model, to phenomenological or physical models that reproduce the exact sensor behavior and raw data streams from interaction with the simulated environment. Such physical sensor models, which are computationally intensive, are typically calculated on high-performance GPU(s) and feed raw data to the connected real sensors, such as camera or radar, for realistic testing in the lab. Achieving close-to-reality sensor models is crucial. This presentation will feature a proposed highly realistic sensor simulation environment and a live demonstration.

Open Standards in Simulation for Testing HAD/AV Systems

Jace Allen, Business Development Manager: Simulation and Test Systems, dSPACE Inc.
Carene Kamel, Business Development Technical Specialist, dSPACE Inc.

Over the years, as simulation became a standard way of testing embedded software, many standards have emerged to create synergies among a myriad of tools available to the automotive engineering community.  dSPACE has been at the forefront of this standardization drive through our active participation in various committees and organizations. SAE, ASAM and recently some research projects like Pegasus have attempted to define standards such OpenDrive, OpenScenario, Open CRG, Open Simulation Interface, FMI, etc. These standards are different than the process standards like ISO 26262 or SOTIF and define the interfaces or formats of data allowing tools to exchange/reuse work artifacts creating efficiencies. However, there is often a dilemma about the comprehensiveness of a standard -- particularly when technologies are evolving.  So what is the current state of standards that will affect the development of autonomous systems? Are we there yet?

TRACK 3 – Validating Autonomous Systems

Validation of Safety - SOTIF & ISO 26262

Jann-Eve Stavesand, Team Leader, Process Consulting, dSPACE GmbH

For the validation of functional safety automotive E/E systems and their embedded software, proper workflows and virtual test methods are necessary. The need for virtual testing and efficient collaboration in the automotive domain is increasing evermore because of new challenges coming from the development of advanced driver assistance systems (ADAS) and autonomous driving (AD). At the same time, the requirements for a reliable process of virtual testing are increasing as ADAS/AD systems are becoming more safety critical.

The presentation outlines necessary development and test strategies that are compliant to ISO 26262 and ISO/PAS 21448 to ensure the proper functionality and safety goals of ADAS/AD functions.

Infusing Reality in ECU Validation: Environment Sensor Interface for HIL Validation

Ben Hager, Project Manager, dSPACE Inc.

To see the surrounding environment, autonomous vehicles are equipped with multiple environment sensors (i.e. camera, LiDAR, Radar, etc.). The data from these sensors has to be processed to validate functions. This presentation spotlights the dSPACE Environment Sensor Interface (ESI) unit and will show how this tool can assist with the simulation of sensors in a HIL setup for sensor fusion and function testing. The ESI unit is able to accurately synchronize the stimulation of individual sensors and supports time-correlated feeding of raw sensor data to one or more sensor ECUs.

Over-the-Air Radar Test Bench for Function Tests and Radar Sensor Evaluation

Dirk Berneck, Group Manager – Rapid Control Prototyping, dSPACE GmbH

Testing radar sensors is one of the elementary challenges when developing advanced driver assistance systems and autonomous cars. This presentation will address:

  • Over-the-air multi-target simulation in a reliable environment.
  • Customer setups and experiences during projects.
  • Where do the echoes come from? Model-based echo generation.