In article 6 of 8 in our Motion Series, Edwin de Vries, CTO at Cruden explains how Cruden works closely with customers to improve upon traditional motion cueing techniques
Simulator motion cueing is where the virtual vehicle’s acceleration, in surge, sway heave, role, pitch and yaw, is translated to fit the driving simulator’s motion system, while taking into account simulator platform workspace, maximum acceleration and maximum velocity. Given the limitations of any motion platform, motion cueing is all about finding the best compromise for specific scenery and experiments.
Often, cueing algorithms induce discrepancies between individual degrees of freedom (DOFs). They can provide false cues (when the platform movement is in the opposite direction to the vehicle movement) and create other issues. The driver’s behavior is influenced by these unwanted factors. The best-known consequence is simulator sickness, although driver sickness is caused by the simulator as a whole; visuals that are not synchronized with the motion can just as easily cause sickness as false motion cues.
Many simulators, and with a variety of hardware configurations, utilize what is commonly known as the Classic Cueing Model. This is a mathematical model in which vehicle acceleration is manipulated using filters and gains to create platform motion to simulator vehicle accelerations while staying within the workspace of the motion platform.
There are issues with the Classic Cueing Model however, which is derived from aerospace use. Cruden has therefore created the Cruden Cueing Model. The aim is to provide motion feedback that is better suited to the driver in every situation, such that the driver behaves more naturally.
Proper motion in a driving simulator begins with a detailed vehicle dynamics model (VDM) and proper configuration of that model. The linear and rotational accelerations generated by the VDM are the input for the Cruden motion cueing algorithms. When we configure a driving simulator, we collaborate with the customer on their vehicle model to tailor its fidelity for use in a driving simulator. This work is done by experienced Cruden vehicle dynamicists and is all about creating a physically correct VDM, involving the road surface, tire model, suspension, correct centre of gravity, inertia and more.
Conventional cueing uses only the vehicle accelerations to determine the motion system’s movement, but Cruden instead takes additional vehicle data from the model, known as a PVA signal (position, velocity, acceleration), while remaining mindful of the available bandwidth to ensure real-time running. This complete and consistent triplet of signals in one specific direction is the foundation for more realistic cueing.
Extracting the PVA signals from a specific position in the virtual vehicle, offset to the left or right depending on where the driver is seated, provides a more natural sensation than trying to calculate the offset from numbers taken from the vehicle’s centre of gravity.
Benefits of cueing approaches are only possible if a rich amount of vehicle data, such as authentic roll and pitch angle, is provided directly by the vehicle to the motion platform.
“Obtaining this information may require not only awareness but also some extra wiring between the model itself and the output interface to the motion base. Our vehicle model integrations are typically based on a MATLAB Simulink interface, so we add some additional lines into the Simulink model. It’s not especially difficult to do, but without it, the cueing results are disappointing. Acceleration alone might have been sufficient in simulators 10 years ago but is no longer enough.”
We’ve seen in a previous article how Cruden reduces washout accelerations and keep them below the human perception threshold. The PVA signals from the VDM open the door to roll or pitch angle cueing as a 1:1 replication of the angle involved in the real maneuver. One example of this would be a simulated vehicle mounting a curb, riding along it at an angle and then dropping back down to the road surface, without the simulator needing to wash out to the horizontal midway through the maneuver. The tilt does not need to be represented by the visual system. Another example would be the option to realistically follow the banking on an oval track or the crown of a road.
The degree to which the motion system is cued to follow the rotation of the vehicle, and the visual system is used to make up the difference, is a matter of customer preference. There is no such thing as general motion cueing. For high-end vehicle dynamics testing, Cruden even customizes and tunes motion cueing algorithms to the specific driver or test scenario.
Different drivers have different human sensory systems. Some drivers demonstrate more realistic driving behavior in the simulator if a larger part of the motion is presented via the visual system, with the motion platform only used for sharp onset cues. By contrast, other drivers like the platform to translate and rotate to the maximum of its capacity.
Our cueing model provides the flexibility to do both. When Cruden builds a simulator, we want to ensure a perfect driving experience and this means we usually support our customers with model validation. This takes away the need for any discussion about a bad driving experience being caused by the customer’s vehicle model.
Our approach seeks to eliminate false cues, and where possible, looks to avoid washout at all, so that the motion supports the visuals with 1:1 roll and pitch angles. For road surface irregularities and undulations, we can also achieve that for heave, but it’s harder with yaw angle, which is limited on a hexapod motion system to around 25 degrees. If true yaw response is important to your simulation, then we suggest adding a yaw table with close to 360 or infinite yaw motion to our motion system, which is the case for two of the Cruden simulators being supplied to BMW.
When it comes to surge and sway motions, our earlier article on workspace also discussed how it is often impossible to replicate these motions 1:1 within a practically sized simulator laboratory. Cruden has nevertheless improved these cues, for example by using a body sideslip measurement to modify the sway cue when the driver turns the steering wheel to drive straight ahead after a long turn.
Surge and sway remain areas where a customer might be better served putting available resources towards other elements of the simulator specification that are more critical to their particular application, rather than spending millions in pursuit of a very large x-y table. In many cases, accepting a slightly smaller scaling or reduction in duration of a surge or sway event is unlikely to disrupt the immersion.
For more information, please contact Dennis Marcus via firstname.lastname@example.org or on +31 20 707 4646.
If you think a colleague would be interested in receiving the articles in this series, you can sign them up to our newsletters here.
Links to subsequent articles will be added below as they are published.
View all articles in our Motion Series of articles: here.
Article 1: Driving Simulator Motion Systems 101