Hazard perception behaviour of motorcyclists in a simulator

• Main Author: Kum, Kevin Yew Kang
• Other Authors: Wong Yiik Diew
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Civil engineering
• Online Access: https://hdl.handle.net/10356/136738
• Institution: Nanyang Technological University

Motorcyclists are often over-represented in the statistics of road traffic fatalities. The proportion of death count involving motorcyclists are higher in the South East Asia region, which results in motorcyclists being termed as the most vulnerable group of road users. Therefore, much research works have been done to identify the causes of these accidents and factors that could reduce their occurrences. One of the factors associated to lower crash rate is good hazard perception skill. Individuals who exhibit better proficiency in hazard perception skill can detect relevant information from their surroundings, assess their current situation and anticipate what are the potential collision courses ahead. In turn, these individuals can take precautionary actions while travelling through hazardous road segments. This study employs the use of a basic motorcycle simulator and eye tracking glasses to study the motorcyclists’ visual search pattern and response time (RT). The simulator was constructed from an abandoned motorcycle frame integrated with simple controls (e.g. manoeuvring, braking and accelerating). Hazardous scenarios were created in a 3D virtual world and were displayed from a viewing screen. This method of testing provides a safe environment and allows active involvement from the test subjects to mimic, to a certain extent, actual riding. Participants were presented with two types of hazard onset occurring at the signalised cross-junction. The first type of hazards is abrupt-onset (hazard which suddenly appears from behind the obstructing object) and the other is gradual-onset (the potential hazard is visible which gradually transforms into a hazard). Next, three groups of subjects (the experienced-E and novice-N riders and the unlicensed-U people) were recruited for this study. In comparing with all RT records, there is significant difference among the hazard onset type (F1,51 = 4.74, p = 0.03) (Wong, 2019) and on average the RT in abrupt-onset is shorter than in gradual-onset by 0.1 seconds. The abrupt-onset hazard may trigger involuntary gaze shift which enhances the process of hazard detection. Thus, this consideration should be considered in future studies that wish to employ the use of abrupt-onset hazard. Upon comparing RT in terms of on-road experiences, there is significant difference among the experience levels (F2,69 = 4.465, p = 0.02) and the unlicensed participants are consistently slower than licensed counterparts (E & N) in both hazard types (Wong, 2019). The licensed subjects also display broader search pattern in abrupt-onset hazard, and they are better at prioritising their gaze in the direction of potential danger for gradual-onset hazard. Thus, the accumulation of on-road experience, in general, does improve one’s HP skills to a certain extent.