Signage such as “Keep Lane Assist On” is Recommended for All Levels of Vehicle Automation, Especially for Lower Levels of Automation.
Simulation of Existing Transportation System on the I-15 Expressway in San Diego County, California Give Insight on the Implications of Adding an AV Exclusive Lane.
Made Public Date
10/26/2021

1173

San Diego
California
United States
TwitterLinkedInFacebook
Identifier
2021-01059

Safety Impact Evaluation of a Narrow-Automated Vehicle-Exclusive Reversible Lane on an Existing Smart Freeway

Background

This project was initiated by Caltrans, looking to increase throughput on four HOV lanes (express lanes) on a seven-mile section of the I-15 in San Diego County. The study evaluated the implications of converting the concrete median that separates the express and regular lanes into a nine-foot automated vehicle (AV) exclusive reversible lane. The study conducted an extensive literature review, an AV manufacturers product review, expert interviews, a consumer questionnaire review, a crash data analysis, and a microscopic traffic simulation analysis. The microscopic simulation model, developed in a commercially available software package, was used to evaluate the impact of the exclusive AV lane on traffic conditions under different market penetration rates (MPR).

Lessons Learned

  • Use highly reflective, clearly visible, and distinct lane markings and signage for proper AV sensor operation. Improve striping on gore and ramp areas to enhance machine vision systems’ performance in AVs. Signage such as “Keep Lane Assist On” are recommended for all levels of automation, especially for lower levels of automation.
  • Implement standardized lane markings, signage, mapping of roads, speed limits, and geometries for AV operations on exclusive lanes.
  • Install physical barriers with active sensors between AV-exclusive and adjacent lanes that are in the same direction to prevent crashes. Lane friction (difference in average speeds of AV-exclusive lane and adjacent lane) could be used to determine if barriers are warranted.
  • Consider nighttime restrictions for AVs. Vision-based sensor systems are vulnerable to low light and wet pavement conditions.
  • Improve all faults in the roadway infrastructure such as potholes and irregular lane markings to reduce the vulnerability of AVs.
  • Widen the lanes around curves, more importantly for the AV-exclusive lane to avoid turning errors. Restrict the operation of heavy AVs if a narrow AV-exclusive lane is implemented for better turning radius and GPS accuracy.
  • Carefully design access points from/to the AV-exclusive lane to allow safe lane changes and manage traffic flow.
  • Allow lane designation to vary from AV-exclusive, shared with HOVs, or open to all traffic depending on the traffic condition and market penetration rate. Expert recommendations suggesting that an MPR of 10 to 30 percent may be sufficient to make AV-exclusive lanes viable.
  • Be aware of AV application limitations. Some AV lane departure warning systems are effective above certain speeds and only work if lane width is between 10 and 15 feet, operate if only two lane markings are detected, and do not perform well in sharp turns, during low visibility, or in foul weather conditions.
  • Educate and train the public at the driver’s license level to explain how the lanes should be used or not used by AV and non-AV drivers.

Safety Impact Evaluation of a Narrow-Automated Vehicle-Exclusive Reversible Lane on an Existing Smart Freeway

Safety Impact Evaluation of a Narrow-Automated Vehicle-Exclusive Reversible Lane on an Existing Smart Freeway
Publication Sort Date
02/01/2021
Author
Machiani, Sahar Ghanipoor; Arash Jahangiri; Benjamin Melendez; Anagha Katthe; Mahdie Hasani; Alidad Ahmadi; and Walter B Musial
Publisher
Prepared by Safety through Disruption (Safe-D) University Transportation Center (UTC) and San Diego State University for USDOT
System Engineering Elements