Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Exploratory Driving Performance and Car-Following Modeling for Autonomous Shuttles Based on Field Data

Renan Favero, Lily Elefteriadou

TL;DR

This study addresses the lack of field-data–driven car-following models for autonomous shuttles by collecting and analyzing trajectory data from AS following a conventional vehicle on public roads in Lake Nona, FL. It evaluates IDM, IIDM, and ACC, calibrating them with a genetic algorithm, and finds that the ACC model provides the most robust fit for AS speed and spacing, while AS trajectories exhibit higher jerk and distinctive acceleration/deceleration patterns. The calibrated AS parameters indicate lower maximum acceleration and higher deceleration relative to AV literature, with IDM and IIDM showing similar behavior and IIDM requiring a lower coolness factor. The work advances pre-field evaluation capabilities for AS, offering a data-driven path to simulate impacts on traffic and guide deployment, with future work extending to gap acceptance, V2I/V2V at intersections, and lane-changing behavior.

Abstract

Autonomous shuttles (AS) operate in several cities and have shown potential to improve the public transport network. However, there is no car following model that is based on field data and allows decision-makers to assess and plan for AS operations. To fill this gap, this study collected field data from AS, analyzed their driving performance, and suggested changes in the AS trajectory model to improve passenger comfort. A sample was collected with more than 4000 seconds of AS following a conventional car. The sample contained GPS positions from both AS and conventional vehicles. Latitude and longitude positions were used to calculate the speed, acceleration, and jerk of the leader and follower. The data analyses indicated that AS have higher jerk values that may impact the passengers comfort. Several existing models were evaluated, and the researchers concluded that the calibrated ACC model resulted in lower errors for AS spacing and speed. The results of the calibration indicate that the AS has lower peak acceleration and higher deceleration than the parameters that were calibrated for autonomous vehicle models in other research

Exploratory Driving Performance and Car-Following Modeling for Autonomous Shuttles Based on Field Data

TL;DR

This study addresses the lack of field-data–driven car-following models for autonomous shuttles by collecting and analyzing trajectory data from AS following a conventional vehicle on public roads in Lake Nona, FL. It evaluates IDM, IIDM, and ACC, calibrating them with a genetic algorithm, and finds that the ACC model provides the most robust fit for AS speed and spacing, while AS trajectories exhibit higher jerk and distinctive acceleration/deceleration patterns. The calibrated AS parameters indicate lower maximum acceleration and higher deceleration relative to AV literature, with IDM and IIDM showing similar behavior and IIDM requiring a lower coolness factor. The work advances pre-field evaluation capabilities for AS, offering a data-driven path to simulate impacts on traffic and guide deployment, with future work extending to gap acceptance, V2I/V2V at intersections, and lane-changing behavior.

Abstract

Autonomous shuttles (AS) operate in several cities and have shown potential to improve the public transport network. However, there is no car following model that is based on field data and allows decision-makers to assess and plan for AS operations. To fill this gap, this study collected field data from AS, analyzed their driving performance, and suggested changes in the AS trajectory model to improve passenger comfort. A sample was collected with more than 4000 seconds of AS following a conventional car. The sample contained GPS positions from both AS and conventional vehicles. Latitude and longitude positions were used to calculate the speed, acceleration, and jerk of the leader and follower. The data analyses indicated that AS have higher jerk values that may impact the passengers comfort. Several existing models were evaluated, and the researchers concluded that the calibrated ACC model resulted in lower errors for AS spacing and speed. The results of the calibration indicate that the AS has lower peak acceleration and higher deceleration than the parameters that were calibrated for autonomous vehicle models in other research
Paper Structure (17 sections, 7 equations, 6 figures, 6 tables)

This paper contains 17 sections, 7 equations, 6 figures, 6 tables.

Table of Contents

  1. Introduction
  2. LITERATURE REVIEW
  3. Existing AV field deployments and their findings
  4. AV movements through models and simulation
  5. Car-Following Models for Autonomous Vehicles (AV)
  6. Car-following Model comparison
  7. METHOD
  8. Data collection
  9. Data cleaning
  10. Trajectory analyses
  11. Model calibration
  12. Model comparison
  13. IDM Car-following Model
  14. ACC Car-following Model
  15. RESULTS
  16. ...and 2 more sections

Figures (6)

  • Figure 1: The first autonomous vehicles in Central Florida operating in Lake Nona
  • Figure 2: This figure shows in green the AS route in the Lake None city, from which it was collected the trajectories data.
  • Figure 3: The process to calibrate the model with Genetic Algorithm (GA)
  • Figure 4: Histogram of AV trajectory variables
  • Figure 5: Speed, acceleration and jerk variability
  • ...and 1 more figures