Joint Attention in Driver-Pedestrian Interaction: from Theory to Practice

Abstract

Today, one of the major challenges that autonomous vehicles are facing is the ability to drive in urban environments. Such a task requires communication between autonomous vehicles and other road users in order to resolve various traffic ambiguities. The interaction between road users is a form of negotiation in which the parties involved have to share their attention regarding a common objective or a goal (e.g. crossing an intersection), and coordinate their actions in order to accomplish it.In this literature review we aim to address the interaction problem between pedestrians and drivers (or vehicles) from joint attention point of view. More specifically, we will discuss the theoretical background behind joint attention, its application to traffic interaction and practical approaches to implementing joint attention for autonomous vehicles.

Looking to Relations for Future Trajectory Forecast

“…This section provides a review of deep learning based trajectory prediction. We refer the readers to [11,17] for a review on recognition and prediction of human action, motion, and intention, and [26,29] for a review on human interaction, behavior understanding, and decision making. Human-human interaction oriented approaches Discovering social interactions between humans has been a mainstream approach to predict future trajectories [31,2,48,23,43,39].…”

Visual attention and its intimate links to spatial cognition

“…Further, Rasouli et al (2018), and others, show that the spatial configuration of an intersection is important (this sounds like it should be obvious, but it is satisfying that it is also supported by data). Crossing behaviour depends strongly on the width of the road, whether it is a designated crosswalk or not, whether there are traffic signals, and more.…”

“…Understanding pedestrians' behaviour at crosswalks is particularly important as they are the most vulnerable road users. Among various factors that influence pedestrian crossing decision (Rasouli & Tsotsos 2018) one crucial component is the configuration of the traffic scene in terms of the locations of pedestrians, other road users, the ego-vehicle and traffic signs and signals. More intuitively, when predicting a pedestrian's behaviour, four factors relating to the spatial configuration of the scene should be considered (see Figure 9): 1.…”

“…The pedestrian's position with respect to the curb (or the crosswalk's boundaries) can indicate whether and where (in the case of intersections) they intend to cross. By standing after the curb, pedestrians often explicitly communicate their intention of crossing to approaching traffic (Rasouli et al 2018). In addition, being close (or belonging) to a larger group of pedestrians may influence the way the pedestrian will behave.…”

“…However, on closer inspection, the visual behaviour of a pedestrian is far from simple or in some sense, standardized. Rasouli et al (2018) show pedestrian motifs extracted from a large video dataset that portray the complexity and variability of actions for a pedestrian when crossing and when deciding to not cross. Pedestrians might use body movements to communicate, look towards the traffic signaling their intention of crossing or demonstrate different walking patterns depending on the situation, e.g.…”