Q&A with Cruden: Simulator for Jaguar Land Rover
Amsterdam-based Cruden designs and manufactures racing and driving simulators. It recently supplied Jaguar Land Rover with a 6-DOF (six degrees of freedom) simulator that is used as a virtual engineering tool by design teams at its Gaydon, UK design and engineering facility. In this interview, Matthew Beecham talked with Cruden's commercial director, Frank Kalff about the simulator for JLR, its benefits and the appetite amongst other OEMs to fine-tune a simulator's motion behaviour to suit their brands.
Could you tell us more about the simulator provided to Jaguar Land Rover?
The Jaguar Land Rover simulator is based on our core Hexatech, 6-DOF full motion simulator. It is extremely versatile, being suitable for both Jaguar and Land Rover vehicles. Jaguar Land Rover will be using it for vehicle dynamics, NVH and ride assessment as well as control systems, aerodynamics, off-road capability and Advanced Product Creation programmes.
The simulator is very realistic and accurate and we have been working with Jaguar Land Rover for some time to customise the interfaces.
On the hardware side, we have integrated an actual production seat, various Jaguar and Land Rover steering wheels (with reach and rake adjustment) and interchangeable pedal boxes which are integrated into the driver cockpit area to provide the most realistic driving position.
On the software side, the simulator runs on Cruden's own Racer Pro operating software. It integrates with the IPG Carmaker vehicle modelling software used by Jaguar Land Rover for its desktop simulation via our ePhyse Net external physics interfacing package. The software allows Jaguar Land Rover to run its vehicle models natively, in co-simulation, with the simulator.
Finally, we have produced LIDAR scanned professional roads and tracks used by Jaguar Land Rover for their virtual testing programme.
Why are simulators becoming so popular?
Simulators are increasingly understood as an alternative to physical pre-prototype building and testing and give vehicle dynamics analysts, who are normally relatively distant to the subjective test drive process, feedback of accurate and realistic 'feel' from their vehicle dynamics models. Simulators can save millions by reducing prototype car production and the costs of going track testing running test rigs. They also open up the possibility to try out many more prototypes compared to testing with actual prototype cars. More settings can be changed and more options evaluated in a simulator than would ever be feasible in the real world, due to budget and time constraints. They are also an instrument for improving corporate social responsibility and reducing the automotive industry's impact on the environment.
Simulators can contribute to the development of components and systems in steering, ride and handling, tyres, HMI and ergonomic assessment, problem investigation, software debugging, competitor assessment and benchmarking and engineer training.
Is it essential for OEMs to have a driving simulator these days?
There a lot of very advanced automotive simulators in use but still not every OEM has one. We find their adoption is very dependent on the culture and acceptance levels of each OEM, both from within the senior management and engineers. Some management teams are vehicle dynamics people at heart; true 'car' people that still get a thrill out driving the cars at test tracks themselves and which give personal input to the development process in the traditional way. Others are very fact-based and business process driven and are therefore much more welcoming of the data that can be generated by a simulator to help drive decision-making.
Saying that, the idea to investigate motion simulation is often driven by a certain engineering team within the OEM, say the steering or NVH team, which recognises its value to achieve a specific goal. Then more people come on board and a simulator development project is born.
Most car manufacturers have some sort of a simulator but the level of sophisticated differs greatly. The way that OEMs solve their problems is different as there are so many different engineering and virtual engineering approaches these days.
Can there be a 'one size fits all' simulator for all vehicle design and development functions to use?
In the OEM automotive engineering space, no. Different applications will require different simulators. For example, for ride and comfort evaluation, you need limited stroke, high frequency, high bandwidth simulation with direct drive actuators. In contrast, ride and handling testing requires much lower (below 20 Hz) bandwidth and larger stroke.
OEMs may have an array of simulators for different things. A central simulator may be used for evaluating handling manoeuvres such as elk tests and lane changes as well as automatic cruise control and driver observation systems. Then there might be a separate one for ride and comfort where engineers experience a ride rather than a drive which can be useful for comparing engine mount A versus B or testing out new dampers. We know of one manufacturer which has a set-up which can test the shifting experience of its gear box on one simulator, for example and the steering on another; it depends on the driving dynamic which is most key to the brand.
Where is simulator technology going next?
Due to the growing expertise among automotive simulator engineers, there is now a much greater appetite for more control over and customization of the simulator's motion behaviour.
This is something we're supporting by allowing our customers to go beyond basic parameter tuning and to be able to custom design algorithms and command direct platform set points from within the Simulink environment.
The increase in computing power is another big factor influencing the world of simulators. So much more can now be done in real time. Together with advancing graphics cards, the visuals are also getting better; in fact they are improving every three to six months or so. Vehicle models can be a lot more complex these days because the calculation power is so good that you can do more in the same time. You can make your car model more detailed and calculate it more accurately and add more parameters.
Surely the test drivers would prefer to drive the real car? Does the simulator involve retraining?
The remainder of this interview is available on just-auto's QUBE research service