The Stuttgart driving simulator
With a diameter of 5.5 m, the dome of the Stuttgart driving simulator provides space for a vehicle mockup of an entire passenger car chassis. Mockups based on existing production vehicles or on non-driving prototypes can be used. Vehicles can be exchanged rapidly, thanks to standardized interfaces in the mockup for mechanical fixation in the dome, power supply and data transfer. The passenger compartment of the vehicle remains completely intact. The control elements of the vehicle function and are assigned different tasks depending on the topic being investigated. Restoring forces from the brakes and steering wheel are reproduced using permanent-magnet synchronous servomotors.
Video recording, eye tracking and position monitoring of the driver's seat are available to assess driver reactions. Simulated vehicle states or states of the assistance systems used are recorded simultaneously to these.
![[Translate to en:] Logo BMBF](/fileadmin/_processed_/0/1/csm_mech_kontext_logo_bmbf_01_da5d79becc.png){kind=logo}
![[Translate to en:] Logo mwkbw](/fileadmin/_processed_/9/e/csm_mech_kontext_logo_mwkbw_01_5923ea4a9d.png){kind=logo}
Visualization system
12 ESP-LWXT-0.6 projectors with WUXGA resolution enable an impressive driving experience. A 241° front projection with approx. 10800x1920 pixels is achieved using 9 projectors. 3 further projectors display the appropriate rearview mirror perspectives on the dome wall. Aided by the openWARP2 System from eyevis, the geometrical corrections required for the curved dome wall are calculated in real-time, within a single frame. Parameterization of the warp matrices is performed using the camera-based auto-calibration system developed by domeprojection.com. The images are rendered by powerful graphics computers with i7-6850K processors and Geforce GTX 1080 TI graphic cards.
Motion system
The motion system of the Stuttgart driving simulator consists of a sled system for simulating large, linear accelerations in the vehicle's longitudinal and lateral directions, plus a hexapod installed on top. Altogether, the system has 8 degrees of freedom and can implement linear movements in a working area of 10 m x 7 m. This means lane-changing and transient operations during starting or gear-changing can be represented without the need for scaling. It is also able to reproduce combined longitudinal and lateral movements which occur both in normal driving and automated driving. It can reproduce accelerations of up to 8 m/s^2 as well as rolling, pitching and yawing movements.
Simulation of vibrations
Using the hexapod, oscillators on the vehicle chassis and a system of loudspeakers, it is possible to cover the entire frequency range of the vibrations occurring in vehicles. This achieves a realistic representation of the ambient noises and vehicle vibrations, as well as focused influencing of tactile and aurally-perceptible vibrations due to road/tire, engine or chassis excitation.
Noise simulation
A variety of loudspeakers are mounted within the dome to reproduce the driving noise conditions. The four loudspeakers positioned on the lower dome wall aid the positioning and representation of surrounding noises, such as passing vehicles. These are also used to generate wind noise and tire rolling noise. The drivetrain noises are simulated by two loudspeakers within the mockup. Two further Bluetooth loudspeakers permit additional noise sources to be positioned at different locations. Digital formats such as MADI and AES3 are used to connect the loudspeakers.
Sensor test bench
Appropriate connections are provided on a separate test bench to incorporate/test (partly) intelligent sensors. Here GPS signals can be provided via a GPS simulation from Spirent, as well as the electronic horizon in ADASIS format. Furthermore, the camera systems can be supplied with appropriate artificial images. Access to the vehicle data is also ensured via a CAN bus. Other communication media such as Ethernet, USB etc. can also be used. A variety of development platforms (VRmagic D3, Nvidia PX2, BP-ECS7800, …) are available for testing new sensor applications.
Connection to other test facilities
The vehicle simulator system is able to communicate with a drivetrain test bench via a real-time-capable data connection. With this combination of "driver-in-the-loop" and "hardware-in-the-loop", the drivetrain receives the driving resistances from the environmental simulation and the driver inputs, such as accelerator position, as inputs. In return, the driver receives direct feedback about the drivetrain behavior. In this way, both sides can profit from an increase in the depth of reality.
Communication network
UDP connections are established to link the different simulation systems. TCP communication enables the configuration of the systems and changing of the parameters during runtime. From December 2017, the integrated Reflective Memory network will, to a great extent, replace the existing UDP communication and enable a rapid, synchronized exchange of data. The configuration data is backed up using an SQL server, permitting repeatability and transferability of the simulations performed.
VALIDATE
As part of the Hi-Tech Strategy and the German government research program IKT2020, the interdisciplinary research project "VALIDATE" was conducted at Stuttgart University from July 2008, with the aim of reducing CO2 emissions. The project was funded by the Federal Ministry of Education and Research until December 2011. The funding enabled the setting-up of a unique research platform to investigate electronic systems for reducing energy consumption, in a partially or completely virtual environment. The Stuttgart driving simulator was constructed as part of this research platform.
It was used firstly to study systems directly affecting energy consumption – through management and control of the powertrain and the vehicle's electrical system. It was also used to study driver assistance systems and information systems which, by supporting the driver during driving, indirectly lead to an energy reduction. Measurement runs in real traffic were able to demonstrate that driving style has a large influence on energy consumption and energy distribution in the vehicle.
Using the research platform, a new driver assistance system was developed which enabled CO2 reductions of 11.3 % in a virtually-simulated, real driving cycle.
ZuSE – Reliability and safety of electric vehicles
In the ZuSE project – funded by the Federal Ministry of Education and Research– research into safety-relevant systems for the entire spectrum of electric vehicles is being conducted in cooperation with ZF Friedrichshafen AG, the Institute for Road and Transport Science and the associated Adam Opel AG. Thanks to the project, the quality of virtual driving tests for electric vehicles using the Stuttgart driving simulator could be improved. Now all electric vehicle-specific characteristics can be simulated, such as, for example, background noise. In combination with the electrified drive system, an integrated development process has been created for testing and validating semi-autonomous assistance functions – as a preliminary to the full automation of road measurements through to volunteer studies in the simulator, which is now being applied in further projects. Furthermore, as part of the project, influencing factors were identified which increase the acceptance of safety-relevant assistance functions.
Determination of the Influence of Drive Noises and Vibrations on Driver Behavior
Funded by the Friedrich und Elisabeth Boysen-Stiftung, the Stuttgart driving simulator was used to investigate the representation of a realistic driving experience. This was achieved by setting up a system which enabled the entire range of vibrations occurring in the vehicle to be reproduced in the driving simulator. Using complex simulation models, a realistic driving experience was determined which was appropriate for the vehicle type and driving condition. Additionally, coupled operation with a multi-configuration powertrain test bench enabled the effect of a real vehicle drive to be reproduced in the driving simulator.
Subjective evaluation of the longitudinal dynamics of different drivetrains
As part of an industrial project in cooperation with Dr. Ing. h.c. F. Porsche AG, the Stuttgart driving simulator was successfully used to subjectively evaluate the longitudinal dynamics of various sports car concepts. The aim of these investigations was to illustrate and evaluate the “drivability” characteristics of digital prototypes as accurately as possible.
The test drivers carried out a variety of tip-in maneuvers on the test track and in the Stuttgart driving simulator. Here, three different sports cars were used. It could be demonstrated that the expert evaluations in the driving simulator and on the test track matched up very well, and that the tip-in maneuvers in the Stuttgart driving simulator were felt to be extremely realistic – both for different gears and speeds as well as for all three sports cars. Another focus of the project was the evaluation of naturally-aspirated and turbo engines. Notably, in this case the differences were recognized significantly more reliably and securely in the Stuttgart driving simulator than on the test track. This is because the tests in the Stuttgart driving simulator are reproducible, and the test subjects can concentrate better on their evaluation task than when on the test track. This makes the Stuttgart driving simulator interesting for future virtual application questions as well.
Subjective evaluation of lateral vehicle dynamics
In the course of a project with Honda R&D Europe (Deutschland) GmbH the Stuttgart Driving Simulator was successfully used for a quantitative evaluation of lateral vehicle dynamics. A first goal was proving that the Stuttgart Driving Simulator is a suitable tool for detailed subjective evaluation of digital prototypes. It was shown that the quantitative effect of various chassis parameters can be assessed and very well correlates to the effect in the real vehicle. The driving simulator can therefore be used for both analyzing and optimizing specific phenomena observed in the real vehicle and for a basic chassis layout using a digital prototype.
In a further step basic research with respect to subjective evaluation of driving comfort and safety at high speed driving was performed. Therefore, representative studies on the impact of different vehicle response signals in different frequency ranges were conducted. Again, clear statement could be made. The results have been taken into account in the development process at Honda R&D Europe (Deutschland) GmbH. Moreover, there is an intensive discussion on current and future fields of application for driving simulation.