In all walks of life, culture can have a major influence on user-behaviour. Automotive engineers have also realised this, developing different versions for global markets. Think of the recent proliferation of long-wheel-base cars for China and how OEMs have focussed efforts on improving the rear seat passenger experience as an example. Such changes can make the difference between choosing one marque or another but what happens when different cultures interface with safety systems ADAS and the rising number of autonomous systems? Do protocols for such systems need to be standardised or should they be customised and how are they best validated? We put these questions and more to Kia Cammaerts, founder of Ansible Motion. Based in Norwich, UK, Ansible Motion provides driving simulator solutions for vehicle constructors and suppliers worldwide.
Firstly, could you explain what your driver-in-the-loop (DIL) simulator offers and how it achieves it?
Go deeper with GlobalData
The aim is to create a fully immersive experience, an environment in which people can interact with a virtual vehicle as if they were interacting with a real one. Thus, it is the human element, a person's emotional and physical involvement, which is the key to creating a compelling DIL simulation. Our simulators can be designed to perfectly replicate an existing car's interior and human-machine interaction space or realise a brand new space from the drawing board. This is not just visual, but tactile as well. The interfaces must feel correct – not just the materials, buttons, switches and touch devices, but the fundamentals must be right. Brake feel, steering torque and so on.
For our high-end simulators, we also offer dynamic motion. We differ from other simulator suppliers that continue to offer hexapods and dart machines. We created a motion base that we call a Stratiform that provides a greater range of overall movements, as well as independent control for a vehicle's primary motion axes. By developing our own motion machinery, we have more direct control over the implementation and seek to create a much more realistic motion experience from the perspective of human perception, so from inside the cabin it feels just as it would in a real car.
If an OEM requests to use your DIL, what does it involve?
We offer a family of static and dynamic simulators that can provide the core for any installation. Each installation, ultimately, is bespoke because customers want to use their simulators in certain ways. We'll spend time with engineers at their facility but also invite them to bring their engineers and test drivers to our R&D centre to optimise the specification for use case today and in the future. Every aspect can be customised around the core architecture. Some customers want to have different cabins, so we have that option. Others are looking for a more generic environment that can be used to validate on-board hardware or software. Because we are all engineers, we spend a lot of time working with them, talking their language. To build a trusted relationship with an OEM or Tier One, we need to prove that we can supply useful tools. I'm very proud of our long-term relationships with our customers.
How well do you really know your competitors?
Access the most comprehensive Company Profiles on the market, powered by GlobalData. Save hours of research. Gain competitive edge.
Thank you!
Your download email will arrive shortly
Not ready to buy yet? Download a free sample
We are confident about the unique quality of our Company Profiles. However, we want you to make the most beneficial decision for your business, so we offer a free sample that you can download by submitting the below form
By GlobalDataWhy would an OEM or Tier One need a fully-dynamic simulator such as yours? Why not use a simpler static or human factors simulator, or even a desktop simulator?
Introducing dynamics into a simulator lab becomes important when the human participant requires a deeper, more realistic immersion experience.
We do build and deploy smaller, stationary simulators as well as our larger, dynamic Delta simulators. For many use cases, a stationary simulator is adequate. Introducing dynamics into a simulator lab becomes important when the human participant requires a deeper, more realistic immersion experience. This usually occurs when the simulation or experiment set involves vehicle trajectory changes at speed. By placing drivers or occupants into simulated environments that replicate as close as possible the real world, the most realistic evaluation results are generated. This helps product development engineers to make accurate decisions. This is quite important when it comes to regional differences as OEMs seek more global solutions.
There has been a lot of talk about semi- and fully-autonomous cars, specifically the change-over process, where control of the car switches from robot to human. How can your simulators help OEMs and Tier Ones as they consider moving from Level 3 to 4?
Our experience is that validation of autonomous systems is one application where feedback from human drivers provides a real impetus for OEMs to rely more heavily on human-in-the-loop simulation technology. A DIL simulator that provides an immersive environment that will 'trick' a participant to behave just as they would in the real world, can provide valuable insight in a safe and controlled environment for how we will react to the novel experience of the intervention of ADAS technologies during a skid, or a collision-avoidance manoeuvre.
Many of these scenarios will be novel to us so if you want to ascertain whether the handover process is going to be confusing helpful, irritating, or dangerous then using a focus group in a DIL lab provides comprehensive objective and subjective data that can't be achieved efficiently at any proving ground. As one example if you want to see how a real person reacts to the car taking over during an emergency, say a child running out in front of the car, DIL simulation now offers that possibility and many millions of others.
With regards to the newer HMI technologies and ADAS systems you just mentioned; what examples have you seen?
We have had very recent first-hand experience of this whereby we have deployed our simulator to assess a range of methods for alerting drivers of an ADAS intervention. Historically, it's fair to say that HMI for these new safety systems has focussed on Western preferences and protocols but as markets have evolved and sales volumes in traditional markets stagnate, that assumption isn't either appropriate or sensible. All OEMs, therefore, need to gain a deeper understanding of regional and cultural differences.
Getting back to the example. One recent use case, where our simulator was placed to support an OEM, was to look at specific ADAS use cases across different continents. From the testing, Japanese drivers expected a different intervention for lane departure warning compared to their US counterparts. There were clear differences in expectations to respond to audible and visual warnings; where and how the messages were displayed for the driver. There are published scientific experiments where certain HMI systems proved unacceptable or confusing for Chinese participants. That extra time to comprehend, particularly in a stressful situation, could be crucial. Other studies show that Chinese drivers are better able to cope with a greater number of simultaneous tasks compared to German or English drivers. It's an area that requires more than just lip service as we move to greater autonomy.
Yet cultural differences have existed long before the autonomous car was dreamed of …
You are right, we have seen cultural differences for a long time and today we still encounter some of these nuances. For example, the indicator stalk on Japanese cars is on the other side to European models. That's one you can quickly adapt to but even if the most rigorous engineer designs an HMI for a new feature in a different market, it's impossible to foresee how cultural factors could impact the usability or even safety of such a system. A system that works well for one culture may be impractical for another. Factor in the higher volume of data presented to drivers and it has the potential to cause issues beyond just frustration.
How does the simulator play a part in this?
As with any aspect of a vehicle, whether that be ride, noise or HMI evaluation, a simulator provides a consistent, repeatable and controlled environment to study and record human interactions with onboard systems. It's more affordable and quicker to make changes than in a real car and it's safe enough to place common drivers in situ and see how they react to scenarios that would be too dangerous to perform even on a controlled proving ground. It's this ability for real people to interact with imagined systems that is key. This is the real motivation for using human-in-the-loop dynamic simulators in the vehicle development process, especially at the early stages. There are other benefits to including human participation in what would otherwise be off-line computer simulations. For example, we can obtain objective biometric data right alongside traditional vehicle data; heart rate, sweat, muscular intention, eye tracking and so on. In the example of lane departure warning alerts, using our simulators in the US and Japan, OEMs are conducting evaluations with local drivers and simply switching configurations and then running new cycles of tests for different markets.
The autonomous car behaving like a 'chauffeur' will largely define the brand experience of the vehicle.
I believe that the autonomous car behaving as a 'chauffeur' will largely define the brand experience of the vehicle. Our perception of the autonomous drive experience and how that fits with brand identity will be important when we decide what service to use. Getting real consumers, in worthwhile sample sizes, to 'experience' this will help OEMs and service providers.
Can you link your simulator to a real sensor to provide an accurate chain of events before and during ADAS intervention?
Yes. And we can go a step beyond. The sensors themselves can be real or simulated. We provide the communication bridges that allow real sensors, operating in Hardware-in-the-Loop test benches, and/or software models of sensors, to be connected directly to our simulator environment. The key here is providing a real-time, shared world where the visual, motion and other sensory content for the human participant are consistent with the world-space content that is recognised by the sensors. Since we're engaging real people, we do not have the luxury of executing our simulations in anything less than real-time, nor can we allow much latency between any of the simulation elements or physical systems. So, we have a rather unique perspective on how to work with sensors for vehicle development environments. The value proposition is quite clear, however. Placing real people into early and often contact with sensor-based ADAS interventions and AI logic for vehicles is required, not only for sorting out the usual product development issues but as a rational step before putting these cars out in the real world.
Since Ansible Motion was founded in 2009, the auto industry megatrends of electrification, autonomous and connectivity have grown in prominence. How has the company addressed these areas?
Yes, we've been trading for a decade now. It feels like it's passed quickly. It's clear that new economic and technical pressures are emerging in the automotive industry, and cars are evolving into something different. On the product development side – our end of things – software and hardware advances naturally flow into this, and we're interested to see to what end. We're certainly seeing more use cases where our driving simulators are being used to evaluate driver assistance systems, V2X connectivity, and advanced electric powertrains. We have developed our simulator to be able to deal with integration with such systems. Integrating sensors, cameras or even entire electric powertrains is possible with our simulators. Our modular approach means that it has been possible to upgrade existing simulators to work with these new systems, without having to buy new hardware.
With regards to greater autonomy, we have to remember the human experience will still be central. If humans remain a part of the equation, a part of determining what is acceptable in the automotive marketplace, then we're confident that our solutions – those that allow real people the opportunity to interact with proposed technologies before they are unleashed into the wild – we will continue to provide some value.
