Driving simulators are extending traditional systems integration capabilities into the digital realm
Vehicle manufacturers are becoming increasingly reliant on component suppliers to contribute their expertise to new products, but it is always down to the car makers themselves to ensure that the different elements are integrated in such a way as to put the vehicle brand’s unique stamp on the finished design.
This could mean adding some finesse to the signature ride and handling behaviour, evolving intuitive HMI (Human Machine Interface) and cabin controls or, increasingly, ensuring ADAS technologies work seamlessly and effectively in order to enhance active safety and either improve or consolidate the brand reputation.
However, as the volume of sensors and software in vehicles increases, it has become more important, and sometimes necessary, to perform this integration process in the virtual world rather than on expensive finished hardware prototypes.
Using software-defined vehicles to overcome the development challenges of integration can result in physical prototypes appearing at a later stage in the programme than was previously the case. Additionally, disciplines in which tuning on the proving ground used to be the norm, such as vehicle dynamics, increasingly turn to driver-in-the-loop (DIL) simulators as a means of proving out their designs and setups before physical cars are available to drive.
A combined approach for added value
Nonetheless, the vehicle manufacturers should not have to abandon their traditional engineering tools just because they’re now testing their vehicles on virtual tracks. Simulator specialists at Cruden believe that a DIL simulator adds value if it can integrate multiple tools in a simple, standard and open way – whether it’s a tyre model, a sensor simulation for ADAS, an acoustics development package or something else. By integrating the engineering tool chain and bringing together different models from different suppliers in the simulator, a driver can experience, and give feedback on, an integrated solution that would otherwise be offered only in a real car.
Control Software
Simulator control software, such as Cruden’s Panthera, has two main purposes. The first is to manage what the driver experiences – the visual cues, motion, steering inputs, audio and more. However, in order to accurately reproduce the vehicle in the virtual world, Panthera’s other function is just as critical: this determines how it integrates with different engineering tools, including vehicle models through a simplified interface based on a software development kit (SDK).
Tyre Models
Tyre models can be used as an example of the integration process. Models have become a required submission from would-be tyre suppliers so that OEM engineers can conduct both offline and online (DIL) simulations and ensure the vehicle’s suspension and tyre perform as a cohesive unit. Often this work extends beyond a Pacejka model for primary ride evaluations into more complex models like FTire to support research into secondary ride, for which it is crucial to communicate higher-frequency inputs from a detailed model of the road surface.
According to Dennis Janssen, Cruden’s Panthera development team lead, the first step is to integrate the tyre model into the OEM’s vehicle model after which Panthera is connected to the vehicle model.
“For that, we have the SDK and ePhyse, our external physics model protocol. Specifically for the road interaction, it might also query SISTer (Server for Interaction with Surfaces & Terrains),” he says.
SISTer determines how the tyre contact patch interacts with the road – either by multisampling with up to 49 intersection queries per wheel, or by driving directly on a dense point set with a spatial density down to 10 mm. SISTer responds to the queries of the vehicle model in less than 1 ms, preserving the immersion of the virtual driving experience. By off-loading the tyre-road interaction from the vehicle model, driving on high-fidelity road surfaces does not require additional computational power from the computer running the vehicle model.
According to Janssen, the SDK-based integration process should take days not weeks and may be accomplished by OEM engineers rather than Cruden specialists, depending on their experience.
“If an OEM engineer is capable in MATLAB Simulink, for example, then they should be able to do it themselves,” he continues. “With the SDK, we take away the detail of how Panthera communicates and how the integration works under the hood. Vehicle model specifics, such as what type of signals need to be available, are in the documentation. To route the signals from your vehicle model to Panthera, you just use the SDK and the functions we provide.”
ADAS Simulation
It’s a similar process to integrate a sensor model for ADAS simulations, whether the goal is to develop a new sensor using a physics-based model – running in co-simulation in a version of the virtual world tailored to the sensor’s input needs – or an ADAS controller and the HMI around it. In the latter case, the simulator can supply ground-truth information to the controller such as lane width and the position of surrounding traffic – information over which it already has full control. “It’s a different type of information being sent, but at a high level, the communication between simulator and models, and the functions we provide to achieve it, are the same,” says Janssen.
Acoustics
Acoustics engineering tools can also be integrated with a driving simulator to facilitate transfer path analysis. Sounds from the OEM’s own prototype or competing cars can be replayed in the simulator and the noise levels tweaked until the acoustics engineers pin down their targets for a new model. Again, the integration process is similar to what has already been seen for other tools.
Because the SDK provides access to every signal living inside Panthera – whether it’s throttle position, engine rpm, vehicle speed, the surface that the car is driving on, or even the position and speed of surrounding vehicles – all of it can be made available to any type of tool. “And once the acoustics engineering tool has been integrated with the simulator, the process does not need to be repeated for each new vehicle being evaluated,” Janssen adds.
Whatever the integration need, Cruden believes its Panthera SDK provides a powerful interface between engineering tools and the heart of the driving simulator. This standardized, rapid approach – including the option to establish FMI interfaces – helps vehicle manufacturers to create an integrated simulation system of their favourite tools. In turn, this enables them to develop better vehicles, faster and at reduced cost.
- UK manufacturing steps up to COVID-19 crisis - April 2, 2020
- Clustering Innovation - March 12, 2020
- A Global Monitor - March 6, 2020