Analysis of Implicit Communication of Motorists and Cyclists in Intersection Using Video and Trajectory Data

Abstract

The interaction of automated vehicles with vulnerable road users is one of the greatest challenges in the development of automated driving functions (ADF). In order to improve efficiency and ensure the safety of mixed traffic, ADF need to understand the intention of vulnerable road users, to adapt to their driving behavior, and to show its intention. However, this communication may occur in an implicit way, meaning they may communicate with vulnerable road users by using dynamic information, such as speed, distance, etc. Therefore, investigating patterns of implicit communication of human drivers with vulnerable road users is relevant for developing ADF. The aim of this study is to identify the patterns of implicit communication of human drivers with vulnerable road users. For this purpose, the interaction between right-turning motorists and crossing cyclists was investigated at a traffic light controlled urban intersection. In the scenario, motorists and cyclists had a green signal at the same time, but cyclist had right-of-way. Using the Application Platform for Intelligent Mobility (AIM) Research Intersection, trajectory and video data were recorded at an intersection in Braunschweig, Germany. Data had been recorded for 4 weeks. Based on the criticality metric post-encroachment time (PET) and quality of the recorded trajectory, 206 cases of interaction were selected for further analyses. According to the video annotation, when approaching the intersection, three common communication patterns were identified: (1) no yield, motorists, who should yield to cyclists, crossed the intersection first while forcing right-of-way; (2) active yield, motorists, who were in front of cyclists, gave the right-of-way; (3) passive yield, motorists, who were behind cyclists, had to give the right-of-way. The analysis of the trajectory data revealed different patterns of changes in time advantage in these three categories. Additionally, the communication patterns were evaluated with regard to frequency of occurrence, efficiency, and safety. The findings of this study may provide knowledge for the implementation of a communication strategy for ADF, contributing to traffic efficiency as well as ensuring safety in the interaction with vulnerable road users.

Communicative Robot Signals: Presenting a New Typology for Human-Robot Interaction

“…Implicit signals are frequently applied to support interactive robots subtly transporting meaning, for example, to request help from a human in a collaborative scenario [3] or by using social proximity to support interaction opening [20]. Implicit signals account for the majority of communication in some situations, such as crossing intersections in traffic [29,59], which stresses the need to identify and characterise them when considering autonomous vehicles as well as HRI. Implicitness of a signal can lead to ambiguity and therefore can provide flexibility and adaptability to the current situations, allowing humans to interpret the exact meaning of the signal (c.f.…”

Interaction Patterns of Motorists and Cyclists at Intersections: Insight from a Vehicle–Bicycle Simulator Study

“…Investigating interaction patterns between VRUs and motorists requires a large amount of real-world data to offset the influence of context factors (e.g., traffic density, time of day) and individual differences (e.g., desired speed). For example, Zhang et al [13] collected the trajectories of over 200 interactions using video-based traffic observation, based on which three interaction patterns were identified: The motorist does not yield to the cyclist, the motorist actively yields to the cyclist, and the motorist passively yields to the cyclist.…”