V&V4NGC

For the safe commissioning of highly and fully automated road vehicles, new model-based methods and toolchains for development, validation, and verification are required. The main focus is on securing and continuously monitoring the behavior of automated vehicles in both known and unknown traffic situations.

Highly automated vehicles
The project focuses on simulation- and scenario-based testing methods.

Methods, Processes, and Toolchains for the Validation & Verification of NGCMethods, Processes, and Toolchains for the Validation & Verification of NGC

The project V&V4NGC (Validation & Verification for the Next Generation Car) is developing scenario-based, simulation-supported procedures that are intended to become a central part of future approval processes for automated driving functions. Continuous improvement is enabled by the use of real field data (e.g., from the Niedersachsen test field). The goal is a seamless lifecycle approach with virtual certification and digital twins, which also takes learning AI systems into account. These approaches are demonstrated and evaluated through a virtual urban drive in Braunschweig.

NGC Simulation Framework

The interaction of automated and autonomous systems with an open world makes it necessary develop completely new approaches for their verification and validation, since their behavior as well as their environment is not capable of being described formally in a complete way. Thus, to bring these societally beneficial systems to the market and into the field in a way that in which they follow our rules and values, it is necessary to explicitly address to prove its quality with methods for their verification and validation, as presented in the following.

The NGC Simulation Framework offers a unique collection of simulation-based methods and tools for the development and testing of highly and/or fully automated (automotive) vehicle systems. The NGC Simulation Framework supports a mixed virtual/physical system integration, enabling key parts of the certification processes for future automation systems and their subsystems in tomorrow's road traffic through a scenario-based simulation environment.

NGC-Simulator
Example elements of the NGC Simulation Framework with custom toolchains on top.

The methods and tools within the NGC Simulation Framework comprise, for example, a macroscopic traffic simulation perspective to model complex traffic environments with a number of different traffic participants (and traffic streams). Below this macroscopic view, different simulation-based tools and methods exists, which can be used for detailed modeling of individual traffic participants as well as components of vehicle automation systems. These in turn are divided into tools and methods for the development of system requirements and vehicle architecture models, up to real-time capable Hardware-in-the-Loop models, which can be used for (virtual) integration tests of HW/SW components of vehicle automations.

Tools and Methods Developed in the V&V4NGC Research Project

Besides the methods and tools for the development and testing of technical systems, another focus is put on methods and simulation-based tools targeting human-machine interaction. This includes real-time capable simulation with human participants in driving simulators, as well as model-based simulation of human-machine interaction using computer models of human behavior. With these different simulation-based methods and tools, a large number of heterogenous studies and tests have been successfully conducted. All of the following reported Verification and Validation (V&V) toolchains use components from the NGC Simulation Framework.

Scenario based verification tool chain

TSC

The TSC toolchain enables efficient simulation-based testing of critical traffic scenarios for the verification of automated driving functions.

RdbS

The RdbS toolchain generates realistic scenarios from vehicle data to improve the simulation of automated driving functions.

Scenario-based Monitoring

Scenario-based online monitoring compares automated vehicle behavior in real time with formal scenario specifications to ensure safe maneuvers and detect novel driving situations.

Safe vehicle dynamics tool chain

AI based Vertical Dynamics Control

Modular toolchain for the development, training, and validation of AI-based vehicle dynamics controllers with direct transfer to real vehicle systems based on established standards such as Modelica and FMI.

Predictive maintenance

AI-based predictive maintenance forecasts the condition of safety-relevant components such as shock absorbers and enables condition-based maintenance to increase reliability and safety.

Digital Twins for Auto-nomous Vehicle Crash Analysis

Simulation-based validation of passive safety structures enables realistic virtual crash tests to enhance vehicle and occupant safety in next-generation mobility concepts.

Safe interaction with road users

Virtual Tester

Model-based HMI simulation with the Virtual Tester enables early evaluation of human-machine interactions and supports efficient, quality-enhancing development of automated systems through realistic behavior simulation.

VRU interaction

The toolchain enables simulation-based testing of the interaction between automated vehicles and vulnerable road users to improve safety and acceptance.

Geospatial Data for Scenario Generation

Digital twins enable simulation-based evaluation and optimization of operational domains and sensor systems of automated vehicles for safe deployment.

Safe & secure online updates tool chain

Modulare Online Updates

Modular online updates ensure interference-free operation of various software components on central control units in the vehicle.

AI-in-the-loop

The AI-in-the-loop toolchain supports real-time training and continuous improvement of safety-critical AI systems.

The V&V4NGC project – Methods, processes and tool chains for validation and verification of the Next Generation Car

Participating DLR institutes and facilities

Institute of Systems Engineering for Future Mobility

The newly founded DLR Institute of Systems Engineering for Future Mobility in Oldenburg focuses on research into the methods needed to develop and validate future automated and autonomous transport systems.

Institute of System Dynamics and Control

The main focus of the Institute of System Dynamics and Control is the requirement driven, system dynamically holistic design, optimal feedforward and robust feedback control of complex safety critical systems for spaceflight systems, aircraft, road, railway and planetary vehicles, as well as industrial robots.

Institute of Space Research

The DLR-Institute of Space Research is a world-leading competence center for optical sensor systems and their application in planetary research, Earth observation and security.

Institute of Transportation Systems

In today’s economy and society mobility is of high significance. People want to arrive at their destinations comfortably and quickly. Goods have to be moved economically over long and short distances.

Institute of Vehicle Concepts

The Institute of Vehicle Concepts is a system-oriented institute, handling and coordinating research subjects related to traffic and transportation at the DLR location in Stuttgart.

Microwaves and Radar Institute

With its know-how and expertise in passive and active microwave remote sensing, the Microwaves and Radar Institute contributes to the development and advancement of ground-based, airborne and spaceborne sensors.

Remote Sensing Technology Institute

Institute of Communications and Navigation

The Institute of Communications and Navigation develops and investigates new systems and methods for radio transmission and radiolocation.

Institute of Flight Systems

The Institute of Flight Systems (FT) has been active in the topics of flight mechanics and measurement and system technology of all flying systems at the DLR site Braunschweig since 1953.

Institute of Software Technology

The mission of the Institute of Software Technology is research and development in software engineering technologies, and the incorporation of these technologies into DLR software projects.

Institute for AI Safety and Security

At the Institute for AI Safety & Security, we research and develop AI-related methods, processes, algorithms, technologies and execution and system environments. Our focus is on safe and standard-compliant AI, cyber security in open data and service ecosystems and AI, as well as autonomous mobility and logistics facilitation.

Contact

Dr. Kim Grüttner

Projectcoordination
German Aerospace Center
Institute of Systems Engineering for Future Mobility
Escherweg 2, 26121 Oldenburg
Germany