Surrogate safety evaluation of incorrect pedestrian detection at signalized intersections

Abstract

The design and operations of signalized intersections are critical for the safety and mobility of all road users. Traditionally, video detection systems have surpassed conventional inductive loops in vehicle detection, offering advantages such as user-friendly interfaces, easy installation, versatile applications, and cost-effective maintenance. Recent technological advancements have expanded the capabilities of video detection systems to include automated pedestrian detection. Pedestrian timing treatment at signalized intersections typically involves two approaches: assuming pedestrian demand during each phase (pedestrian recall operations) or activating pedestrian demand when a button is pressed. Automated Pedestrian Video Detection Systems (APVDS) represent a newer approach where pedestrian calls are automatically initiated upon detecting pedestrians within the detection zone. Both pedestrian push buttons and APVDSs have their pros and cons. While push buttons can cause confusion and incorrect usage among pedestrians, APVDSs theoretically address these issues by autonomously determining pedestrian phase needs. However, challenges arise due to differing pedestrian behaviors compared to vehicles, potentially leading to missed or false calls and unsafe pedestrian actions. This study examines the impact of two APVDS issues: missed pedestrian calls and false pedestrian calls. It also compares the effects of incorrect calls under various operational strategies, including APVDS with and without controller pedestrian recycle, pedestrian recall operations, and push-button operations, with and without pedestrian recycle, across different times of the day to replicate varying vehicular and pedestrian demands. Using Vissim microsimulation software and Surrogate Safety Assessment Model (SSAM), experiments were conducted to replicate field conditions closely. This microsimulation approach enables the measurement and analysis of missed and false calls' impact on surrogate safety measures, providing valuable insights challenging to obtain in real-world conditions. The results contribute to pedestrian traffic signal strategies and enhancing pedestrian safety at signalized intersections in urban environments.

Keywords

• detection
• pedestrians
• surrogate measures of safety (SMoS)
• traffic signals