Utilizing LiDAR Sensors to Detect Pedestrian Movements at Signalized Intersections
-
2022-12-01
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final Report Sept 2021 to December 2022
-
Corporate Publisher:
-
Abstract:National Highway Traffic Safety Administration (NHTSA) has reported that pedestrian fatalities increased by 44% from 2010 to 2019, and more than 20% of pedestrian fatalities occurred at intersections. To improve pedestrian safety, it would be necessary to first observe evolving pedestrian behaviors. To serve this purpose, in this research project, the investigators of the University of Utah and the University of Texas at Arlington jointly explored state-of-the-art LiDAR technology to detect and track vehicles and pedestrians in real time at signalized intersections. Compared with general LiDAR sensing technologies, the investigator has developed application-specific algorithms on top of generic perception algorithms to collect pedestrian behaviors. The developed algorithm can synchronize and fuse two types of real-time information: tracked pedestrians (by LiDAR) and real-time traffic signal status. In this context, three pedestrian behaviors were collected as a proof of concept: (1) Waiting time before they entered the intersection; (2) Effective perception-reaction time to the onset of WALK, and (3) crossing speed. Other than the proposed data collection, the research team also evaluated the scalability and potential training efforts for deployment at scale.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type: