Article Preview
Top1. Introduction
Driving simulators were born during the early 1930s as a tool to safely evaluate the driving skills of a dangerously growing pool of reckless automobile owners in the UK (Fisher, 2011). Since then, the use of driving simulators has been often tied to minimize the danger that untrained drivers pose to the public, as was the case in Japan in the 1960s (Yoshimoto, 2008). Inherently driving simulators are instruments that demonstrate realistic driving experiences in controlled environments and researchers today still utilize them primarily to evaluate traffic safety. However, simulators have evolved to be more than a mere tool for training and education.
Automakers have purposely extended simulator capabilities to evaluate the stability, maneuverability and ergonomics of vehicle design concepts during the holistic development stage. By doing this, the driving simulator has become a tool for innovation. Thanks to the continuous increase in levels of realism of driver user experiences, researchers can have at their disposal a wide range of driving simulator configurations, from low to high fidelity across visual, kinesthetic and auditory stimuli (Mueller, 2014). Even the end-customer now has access to low budget driving simulators with highly immersive user experiences, since realistic, low-cost, 3D virtual environments have been applied to race car gaming (Backlund, 2009).
Depending on research goals, development phase, fidelity requirements and available budget, an automotive user experience researcher could decide to use a minimalistic paper prototype IVI (In-Vehicle Infotainment) mockup or run a series of studies in high-end motion-feedback-equipped simulators. The benefits and drawbacks of each setup have been widely debated in Weinberg 2009, Boyle 2010, and Chan 2010. The reality is that the majority of the researchers compromise with mid-fidelity driving simulators, especially those whose physical configuration can integrate with open access software. Therefore, in the recent years we have seen a proliferation of academic papers making use of open source driving simulator software such as OpenDS (Math, 2012) or the Lane Change Task tool (Mattes, 2003). The success of these tools is tightly related to the ability to perform standardized measures of mental workload, driver performance or subjective evaluations as advocated by Green (2012, 2013). However, these platforms, while great at measurement, still lack the flexibility to become a development framework.
An HMI (Human-Machine Interface) concept that needs to be evaluated in driving conditions requires an investment and effort akin to product development on current platforms. This contradicts the dimensional nature of a designer’s work. As Hendrie (2015) states, “innovation does not begin with a single version but rather an experimentation that moves from one form of output to another, raising issues and communicating in an iterative / generative design process.” Therefore, we wanted to create a driving simulator platform that unleashed creativity and enabled design thinking practices through making. We named it Skyline. The rest of this paper is targeted to provide an overview of the Skyline platform and is structured as follows. In section 2 we provide background on technical components of Skyline, including the physical and software components. In section 3 we further explain the design principles that Skyline enables and in section 4 we describe the practical application of Skyline presenting a case study of control vs automation in the vehicle. Finally, Section 5 summarizes the conclusion and outlines future work.