When going somewhere new, what type of transportation do we take to reach our destination in an efficient and economic manner, and will our choice be different from our daily commute? When walking on the street, how do we perceive and react to someone riding an e-scooter towards us?
At the NTU-LTA Transport Research Centre (NTU-LTA TRC), a partnership between NTU and Singapore’s Land Transport Authority (LTA), a multidisciplinary team, including researchers from NTU’s Schools of Art, Design and Media, Civil and Environmental Engineering, Mechanical and Aerospace Engineering, and Social Sciences, is working on projects to improve mobility efficiency and enhance transport safety on the road.
The use of virtual reality (VR) allows researchers to systematically manipulate the layout of the environment and design different kinds of experimental conditions. Here at NTU-LTA TRC, we address questions in transport research from the user’s perspective, by simulating and modifying the environment in VR. Using a variety of VR test scenarios, our research team studies the perception, behaviour and response of pedestrians, cyclists and motorists in walking, cycling and driving simulators developed by the team.
Walking simulators show the way
In our wayfinding project, we investigate the impact of spatial elements on human perception, so as to provide recommendations for effective infrastructure design at Singapore’s mass rapid transit (MRT) stations.
First, VR models are generated based on a 3D model created by Singapore’s Urban Redevelopment Authority, using the 3D virtual software UC-win/Road (Figure 1).
Subjects are asked to wear VR equipment, such as the HTC Vive headset with a built-in eye tracker, and to “walk” towards several target destinations using a walking simulator (Figure 2). Since the VR scene is synchronised with the subject’s walking behaviour, any changes in speed and direction on the simulator are reflected in the speed of movement and view of the VR scene.
Using these scenarios, we test factors such as the width between platform screen doors and escalators and between escalators and fare gates, the placement of signage and artwork, colour schemes, and ambient lighting on subjects’ behaviour. By recording and analysing our subjects’ response accuracy, reaction time and eye movement patterns, we can compare various modifications of the VR environment and study the impact of colour coding of signs or the addition of signage at critical locations.
Sharing footpaths safely
To better understand how pedestrians and cyclists share pathways, we are studying human behaviour with the help of a virtual cycling simulator platform projected through an HTC Vive head-mounted display. This specially designed cycling simulator allows us to synchronise cycling direction and speed, including the effects of braking and shifting of gears, with VR scenarios (Figure 3).
Analysing reaction time, speed perception and the general behaviour of both cyclists and pedestrians in simulated scenarios, we test whether the current speed limit of 15km/h on footpaths is safe for users. We also hope to shed light on cycling and walking habits in Singapore, such as whether cyclists and pedestrians keep to the left, whether cyclists slow down near bus stops or obstructed-view bends, and how people react to scenarios such as the sudden appearance of other path users or cyclists overtaking them. In addition, we analyse eye-movement patterns and the behavioural responses of cyclists and pedestrians on pavements to find out whether the positioning of signage is effective.
One for the road
Our team at NTU-LTA TRC also designs virtual environments to test the perception and behavioural responses of car drivers to different information delivery modes of in-vehicle units (Figure 4).
To enhance safe and effective interactions of drivers with information displays and graphical user interfaces, we study the attention, reaction and preferences of drivers in a driving simulator to find out when they read traffic information or emergency messages from in-vehicle units (Figure 5). Based on the data, we hope to improve current systems such that drivers receive traffic alerts promptly, while ensuring their attention is not diverted too frequently from the road.
Continuous innovation of user-centred infrastructure is needed to meet the mobility needs of modern cities. These needs might include dedicated cycling paths for cyclists and users of personal mobility devices such as e-scooters away from pedestrians, efficient transition systems between different modes of public transport, or timely information on road congestion and parking spots for car drivers. By shedding light on the behaviour of different users in various transportation scenarios, NTU-LTA TRC’s work drives the building of tomorrow’s smart mobility infrastructure.