Falls are responsible for a large proportion of serious injuries and deaths among cyclists [1-4]. A common fall scenario is loss of balance during an emergency braking maneuver to avoid another vehicle [5-7]. Automated Vehicles (AV) have the potential to prevent these critical scenarios between bicycle and cars. However, current Threat Assessment Algorithms (TAA) used by AVs only consider collision avoidance to decide upon safe gaps and decelerations when interacting wih cyclists and do not consider bicycle specific balance-related constraints. To date, no studies have addressed this risk of falls in safety critical scenarios. Yet, given the bicycle dynamics, we hypothesized that the existing TAA may be inaccurate in predicting the threat of cyclist falls and misclassify unsafe interactions. To test this hypothesis, this study developed a simple Newtonian mechanics-based model that calculates the performance of two existing TAAs in four critical scenarios with two road conditions. Tue four scenarios are: (1) a crossing scenario and a bicycle following lead car scenario in which the car either (2) suddenly braked, (3) halted or (4) accelerated from standstill. These scenarios have been identified by bicycle-car conflict studies as common scenarios where the car driver elicits an emergency braking response of the cyclist [8-11] and are illustrated in Figure 1. The two TAAs are Time-to-Collision (TTC) and Headway (H). These TAAs are commonly used by AVs in the four critical scenarios that will be modelled. The two road conditions are a flat dry road and also a downhill wet road, which serves as a worst-case condition for loss of balance during emergency braking [12].

Drivers continually interact with other road users and use information from the road environment to make decisions to control their vehicle. A clear understanding of different parameters impacting this interaction can provide us with a new design approach for a more effective driver assistance system - a personalised trajectory prediction system. This paper highlights the influential factors on trajectory prediction system performance by (i) identifying driver behaviours impacting the trajectory prediction system; and (ii) analysing other contributing factors such as traffic density, secondary task, gender and age group. To explore the most influential contributing factors, we first train an interaction-aware trajectory prediction system using time-series data derived from the Second Strategic Highway Research Program (SHRP2) Naturalistic Driving Study (NDS). Prediction error is then analysed based on driver characteristics such as driver profile which is subjectively measured through self-reported questions, and driving performance which is based on evaluation of time-series information such as speed, acceleration, jerk, time, and space headway. The results show that prediction error significantly increased in the scenarios where the driver engaged in risky behaviour. Analysis shows that trajectory prediction system performance is also affected by factors such as traffic density, engagement in secondary tasks, driver gender and age group. We show that the driver profile, which is subjectively measured using self-reported questionnaires, is not as significant as the driving performance information, which is objectively measured and extracted during each specific driving scenario.