The I-STREET Testbed

March 31, 2021 in Centers, Research, Technology Transfer News, Testbed, Uncategorized

The information contained on this page is currently being used for the 2021 CoMotion Miami Challenge Powered by I-STREET.

“I-STREET” is a real-world testbed located in Gainesville, Florida. The testbed is dedicated to advancing the state-of-the art in connected and automated vehicle implementation, and it is a joint effort between the Florida Department of Transportation, the University of Florida Transportation Institute (UFTI), the City of Gainesville, and industry partners. I-STREET addresses V2X implementation for vehicles, pedestrians, and bicycles; it utilizes machine learning and AI; and it uses a database platform to assess safety and mobility across the testbed. With the confluence of UF, Gainesville, and Interstate 75, I-STREET leverages a multimodal roadway network and emerging transportation technologies to advance safety, mobility, workforce, and innovation.

I-STREET Infrastructure

I-STREET testbed contains the following components:

  • 129 Road Side Units (RSUs)
  • 56 On Board Units (OBUs)
  • 5 Autonomous Vehicles (AVs)
    • Gainesville Autonomous Shuttles (4)
    • Research AV – NaviGATOR (1)
  • Custom Instrumentation from Research at UFTI
    • DSRC Communication Equipment (Cohda Mk5 radios, Mobilemark Antennas)
    • Smartmicro 2D and 3D Radars
    • Econolite Autoscope Vision camera
    • Real-time Intersection Optimizer (RIO) Software Program
  • Other Equipment
    • Mobileye Shield+ – Advanced Driver Assistance System (ADAS) on RTS Transit buses
    • Bike Rack Sensors on RTS transit buses
    • Eye Trackers
    • Smartphone Applications

I-STREET Data

Signal Data

Format: [Data sets], (B) type, (C) bin rate/periodic refresh rate, (D) granularity (E) method of acquiring

  • [ATSPM data]: (B) tabular, (C) bins are every 5 minutes, (D) contents every 0.1 seconds, (E) via API through spm.traffops.com,  or directly from controller FTP/SCP protocol
  • [DSRC SPaT data]: (B) tabular, (C and D) real-time, every 0.1 seconds, (E) via API through “Siemens” XFER protocol or directly from OBU that is enrolled in SCMS
  • [Signal timing and Coordination data]— (B) Mostly static, unless we (or a consultant) retime an arterial segment. (C and D) only when changed (E) via Trafficware’s ATMS.now client

Traffic Count/Flow Data

  • Detection Data (B) Tabular (C) bins every 24 hours; (D) every 15 minutes (E) via  GridSmart Client; or Iteris Client This can be a per approach (i.e. like a turning movement count); or per detector channel (i.e. say per lane of a 3 lane approach)
  • Arterial Segment Travel Time Data
    • Trapezium has the Siemens/Acyclica (uses Bluetooth and Wifi temporal address matching)
    • FRAME (several arterials – about 60+ intersections) has the TrafficCast/Bluetooth Data
    • Waze for Cities (but not very granular)  —our waze for Cities feed is now link within our “TrafOps” portal
  • ADT/Average Daily Traffic or Tube counts — 90 count locations are obtained every other year, during the Spring and fall

Video Data (note City of Gainesville Traffic Operations does not record, they can share the stream to researchers)

  • Fisheye camera from GridSmart Video Detection(via VLC or other open source RTSP client) about 35 locations
  • Fixed camera from Iteris Video Detection (via VLC or other open source RTSP client) – about 50 locations
  • Pan-Tilt-Zoom (PTZ) camera (CCTV Devices from Bosch) – (via VLC or other open source RTSP client) – about 200 locations.