Tracking and Simulating Pedestrian Movements at Intersections Using Unmanned Aerial Vehicles

Abstract

For a city to be livable and walkable is the ultimate goal of future cities. However, conflicts among pedestrians, vehicles, and cyclists at traffic intersections are becoming severe in high-density urban transportation areas, especially in China. Correspondingly, the transit time at intersections is becoming prolonged, and pedestrian safety is becoming endangered. Simulating pedestrian movements at complex traffic intersections is necessary to optimize the traffic organization. We propose an unmanned aerial vehicle (UAV)-based method for tracking and simulating pedestrian movements at intersections. Specifically, high-resolution videos acquired by a UAV are used to recognize and position moving targets, including pedestrians, cyclists, and vehicles, using the convolutional neural network. An improved social force-based motion model is proposed, considering the conflicts among pedestrians, cyclists, and vehicles. In addition, maximum likelihood estimation is performed to calibrate an improved social force model. UAV videos of intersections in Shenzhen are analyzed to demonstrate the performance of the presented approach. The results demonstrate that the proposed social force-based motion model can effectively simulate the movement of pedestrians and cyclists at road intersections. The presented approach provides an alternative method to track and simulate pedestrian movements, thus benefitting the organization of pedestrian flow and traffic signals controlling the intersections.

UAVs for traffic monitoring: A sequential game-based computation offloading/sharing approach

“…For this purpose, an analytical methodology has been presented for analyzing traffic flow conditions at signalized intersections. Zhu et al [21] propose a drone-based method to detect and simulate pedestrian movements at intersections with frequent collisions. Using indepth learning methods and high-resolution extraction of pedestrian, bicycle and vehicle movements, they were able to calibrate a model to extract pedestrian trajectories and study them to protect pedestrian lives.…”

“…Nowadays, with a considerable increase in the number of vehicles, road traffic monitoring and management is becoming a major challenge for the authority. To overcome this problem, some works [8][9][10][11][13][14][15][16][17][18][19][20][21][22][23][24] propose to equip authority vehicles, such as police and emergency vehicles, with UAVs to assist them in traffic monitoring operations. UAVs can detect vehicles that have committed traffic offenses such as speeding or crossing a continuous line [27].…”

Evaluating the Impact of Drone Signaling in Crosswalk Scenario

“…In Reference [31], authors tackled pedestrian safety diagnosis at signalized crosswalks using traffic conflict technique, and emphasized that the most serious pedestrian-vehicle conflicts occur when vehicles travel at high speed. Tracking and simulating pedestrian movements is part of the solution to protect their safety at intersections [7]. In Reference [32], the reader can also find models for pedestrian path prediction at 2 s time horizon, and this helps to analyze interactions between pedestrians and straight-going vehicle at non-signalized crosswalks.…”

“…Despite this restrictive legal context, various recent works have shown the benefits of UAV technologies to improve efficiency and safety in urban areas. In most of these applications, UAVs have a certain level of intelligence, which allows them to be used as high-performance sensors, information collectors, or even as communication relays, particularly when land cover is not sufficient [7,8]. The advances of C-ITS technologies and the interesting properties of UAVs provide a favorable context to introduce a UAV in a use case where we consider a conflict involving a car and a pedestrian.…”

How Do Human-Driven Vehicles Avoid Pedestrians in Interactive Environments? A Naturalistic Driving Study

“…Although the NDS has the disadvantage of efficiency and security compared to the other two methods due to the accidental and uncontrollable interaction environments, it can reflect the real behaviour of road users, which cannot be affected by experiment conditions and related deviations [ 16 ]. Pedestrian and vehicle behaviour data acquisition, e.g., pedestrian position, can often be carried out through fixed videography, in-motion videography, or other sensors [ 17 , 18 , 19 , 20 ], and most current studies focus on predicting whether pedestrians cross the street based on these data [ 21 , 22 , 23 ]. It is known that the pedestrian crossing intention can change at any time, and the driver’s subjective judgement on pedestrian crossing intention (SJPCI) can affect vehicle deceleration behaviour directly.…”