Surrogate Safety Measures Prediction at Multiple Timescales in V2P Conflicts Based on Gated Recurrent Unit

Abstract

Improving pedestrian safety at urban intersections requires intelligent systems that should not only understand the actual vehicle–pedestrian (V2P) interaction state but also proactively anticipate the event’s future severity pattern. This paper presents a Gated Recurrent Unit-based system that aims to predict, up to 3 s ahead in time, the severity level of V2P encounters, depending on the current scene representation drawn from on-board radars’ data. A car-driving simulator experiment has been designed to collect sequential mobility features on a cohort of 65 licensed university students who faced different V2P conflicts on a planned urban route. To accurately describe the pedestrian safety condition during the encounter process, a combination of surrogate safety indicators, namely TAdv (Time Advantage) and T2 (Nearness of the Encroachment), are considered for modeling. Due to the nature of these indicators, multiple recurrent neural networks are trained to separately predict T2 continuous values and TAdv categories. Afterwards, their predictions are exploited to label serious conflict interactions. As a comparison, an additional Gated Recurrent Unit (GRU) neural network is developed to directly predict the severity level of inner-city encounters. The latter neural model reaches the best performance on the test set, scoring a recall value of 0.899. Based on selected threshold values, the presented models can be used to label pedestrians near accident events and to enhance existing intelligent driving systems.

Before-after safety analysis of a shared space implementation

“…Low values of TAdv are indicative of a potentially risky situation, and TAdv = 0 means that the two users are on a collision course. Recent examples of its application in VRU safety research exist (Gruden et al, 2022;Miani et al, 2022Miani et al, , 2021, but also this measure has a relevant shortcoming. As explained by Laureshyn et al (2010), TAdv "is by itself not sufficient to describe the collision risk since it is also important to know how soon the encroachment will occur"; in practice, even if at time t TAdv is small, if the expected encroachment is far away, the users would have plenty of time to adjust their speed/trajectories and avoid each other without significant risk.…”

“…As TTC, during a traffic encounter, TTAC is measured continuously and the minimum value of TTAC is usually considered as a single representative indicator of its severity. Despite this, only very few works applied TTAC for road safety evaluations (Borsos, 2021;Borsos et al, 2020;Miani et al, 2021). More details on its calculation are given in Section 3.2.…”