Bigdata Analytics and Artificial Intelligence for Smart Intersections

Details

• Creators:
  Banerjee, Tania ; Chen, Ke ; Huang, Xiaohui ; He, Pan ; Emami, Patrick ; Wu, Aotian ; Elefteriadou, Lily ; Rangarajan, Anand ; Srinivasan, Siva ; Ranka, Sanjay
• Corporate Creators:
  University of Florida Transportation Institute ; University of Florida. College of Engineering
• Corporate Contributors:
  Florida Department of Transportation
• Subject/TRT Terms:
  • Artificial intelligence
  • Cloud computing
  • Data analysis
  • Data collection
  • Infrared detectors
  • Intersections
  • Laser radar
  • Pedestrian movement
  • Pedestrians
  • Vehicle trajectories
  More +
• Resource Type:
  Tech Report
• Geographical Coverage:
  United States
• Edition:
  Final Report 07/03/2019 - 06/30/2022
• Corporate Publisher:
  Florida Department of Transportation
• Abstract:
  This project aims to develop a multi-sensor system for vehicle and pedestrian traffic analysis at traffic intersections. The techniques developed as part of this project are used to process data streams from video-camera and LIDAR systems installed at traffic intersections (along with the loop detector data captured by advanced traffic controllers). We use state-of-the-art computer vision and machine learning to perform vehicle tracking (localization, tracking, turn estimation) and pedestrian monitoring from data streams obtained using standard fisheye cameras and LIDARs mounted at intersections and loop detectors beneath the streets. This tracking information is then used to derive motion patterns or trajectories. The trajectories can be utilized for traffic conflict detection and safety analysis and to improve signal timing plans for increased traffic throughput. LIDAR data is beneficial for counting the pedestrians and bicyclists and tracking the patterns of movement they follow while crossing the intersection especially when lighting conditions are poor.
• Format:
  PDF
• Funding:
  BDV31-977-116
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/75114/dot_75114_DS1.pdf
• File Type:
  [PDF - 10.46 MB]
• Collection(s):
  Artificial Intelligence Deployment Center of Excellence (AI COE) ; US Transportation Collection
• Main Document Checksum:
  urn:sha-512:be3f21a9be18a9afd614fa6ac415ba4397a0552e74ae31fbcb69e4ca274bb7fe7fc763127a0f3f72ef573bf843c40215e0e9983d59ade6e5068f4191b80b67ea