Trajectory Dataset for translation of driver-pedestrian behavioral models at semi- controlled crosswalks into a quantitative framework.

<p>The research described in this project began in response to frequent questions from users of several crosswalks near a university campus.  At each crosswalk was a sign indicating that motorists should yield to pedestrians in the crosswalk.  That this message was not being interpreted uniformly was a concern at locations where heterogeneous road users (pedestrians, cyclists, and motorists) were interacting.  Instead of trying to impose a single interpretation on users of each crosswalk, it was decided to observe and analyze interactions between users of the crosswalk. </p>

<p>Recently, computer vision (CV) algorithms have been extensively used in road users’ detection and tracking at an unparalleled spatial-temporal scale. In this study, CV algorithms have been applied to convert the video recordings into a large-scale spatial-temporal trajectory dataset including 800 pedestrians and cyclists interacting with more than 500 vehicles.</p>

<p>Trajectory profiles in this dataset can be used to evaluate behaviors of multiple road users when they interact in uncontrolled shared spaces. Similar to the applications of KITTI, ApolloScape, BDD100K, and Argoverse, the open-sourced dataset can be used in pedestrian planning, simulation, and prediction tasks. Detailed applications of the trajectory/interaction dataset can be found in the dissertation: </p>

<p>Zhang, Y. (2022). <i>Making Crosswalks Smarter: Using Sensors and Learning Algorithms to Safeguard Heterogeneous Road Users</i> (Doctoral dissertation, Purdue University Graduate School).</p>

<p dir="auto">Each folder contains trajectory profiles of pedestrians (*.csv starts with "<strong>ped</strong>") and trajectory profiles of motorists (*.csv starts with "<strong>veh</strong>"). Pedestrian trajectory profiles for Time 4 are now being pre-processed currently and will be updated later.</p>