Real-Time Safety Optimization of Connected Vehicle Trajectories Using Reinforcement Learning
Speed advisories are used on highways to inform vehicles of upcoming changes in traffic conditions and apply a variable speed limit to reduce traffic conflicts and delays. This study applies a similar concept to intersections with respect to connected vehicles to provide dynamic speed advisories in...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-06-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/21/11/3864 |
_version_ | 1797531467471912960 |
---|---|
author | Tarek Ghoul Tarek Sayed |
author_facet | Tarek Ghoul Tarek Sayed |
author_sort | Tarek Ghoul |
collection | DOAJ |
description | Speed advisories are used on highways to inform vehicles of upcoming changes in traffic conditions and apply a variable speed limit to reduce traffic conflicts and delays. This study applies a similar concept to intersections with respect to connected vehicles to provide dynamic speed advisories in real-time that guide vehicles towards an optimum speed. Real-time safety evaluation models for signalized intersections that depend on dynamic traffic parameters such as traffic volume and shock wave characteristics were used for this purpose. The proposed algorithm incorporates a rule-based approach alongside a Deep Deterministic Policy Gradient reinforcement learning technique (DDPG) to assign ideal speeds for connected vehicles at intersections and improve safety. The system was tested on two intersections using real-world data and yielded an average reduction in traffic conflicts ranging from 9% to 23%. Further analysis was performed to show that the algorithm yields tangible results even at lower market penetration rates (MPR). The algorithm was tested on the same intersection with different traffic volume conditions as well as on another intersection with different physical constraints and characteristics. The proposed algorithm provides a low-cost approach that is not computationally intensive and works towards optimizing for safety by reducing rear-end traffic conflicts. |
first_indexed | 2024-03-10T10:44:13Z |
format | Article |
id | doaj.art-bcdf3460f11b416791a3a8e6caf48aba |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-10T10:44:13Z |
publishDate | 2021-06-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-bcdf3460f11b416791a3a8e6caf48aba2023-11-21T22:42:38ZengMDPI AGSensors1424-82202021-06-012111386410.3390/s21113864Real-Time Safety Optimization of Connected Vehicle Trajectories Using Reinforcement LearningTarek Ghoul0Tarek Sayed1Department of Civil Engineering, University of British Columbia, 6250 Applied Science Lane, Vancouver, BC V6T 1Z4, CanadaDepartment of Civil Engineering, University of British Columbia, 6250 Applied Science Lane, Vancouver, BC V6T 1Z4, CanadaSpeed advisories are used on highways to inform vehicles of upcoming changes in traffic conditions and apply a variable speed limit to reduce traffic conflicts and delays. This study applies a similar concept to intersections with respect to connected vehicles to provide dynamic speed advisories in real-time that guide vehicles towards an optimum speed. Real-time safety evaluation models for signalized intersections that depend on dynamic traffic parameters such as traffic volume and shock wave characteristics were used for this purpose. The proposed algorithm incorporates a rule-based approach alongside a Deep Deterministic Policy Gradient reinforcement learning technique (DDPG) to assign ideal speeds for connected vehicles at intersections and improve safety. The system was tested on two intersections using real-world data and yielded an average reduction in traffic conflicts ranging from 9% to 23%. Further analysis was performed to show that the algorithm yields tangible results even at lower market penetration rates (MPR). The algorithm was tested on the same intersection with different traffic volume conditions as well as on another intersection with different physical constraints and characteristics. The proposed algorithm provides a low-cost approach that is not computationally intensive and works towards optimizing for safety by reducing rear-end traffic conflicts.https://www.mdpi.com/1424-8220/21/11/3864trajectory optimizationdynamic speed advisoryreal-time safetyconnected vehiclesreinforcement learning |
spellingShingle | Tarek Ghoul Tarek Sayed Real-Time Safety Optimization of Connected Vehicle Trajectories Using Reinforcement Learning Sensors trajectory optimization dynamic speed advisory real-time safety connected vehicles reinforcement learning |
title | Real-Time Safety Optimization of Connected Vehicle Trajectories Using Reinforcement Learning |
title_full | Real-Time Safety Optimization of Connected Vehicle Trajectories Using Reinforcement Learning |
title_fullStr | Real-Time Safety Optimization of Connected Vehicle Trajectories Using Reinforcement Learning |
title_full_unstemmed | Real-Time Safety Optimization of Connected Vehicle Trajectories Using Reinforcement Learning |
title_short | Real-Time Safety Optimization of Connected Vehicle Trajectories Using Reinforcement Learning |
title_sort | real time safety optimization of connected vehicle trajectories using reinforcement learning |
topic | trajectory optimization dynamic speed advisory real-time safety connected vehicles reinforcement learning |
url | https://www.mdpi.com/1424-8220/21/11/3864 |
work_keys_str_mv | AT tarekghoul realtimesafetyoptimizationofconnectedvehicletrajectoriesusingreinforcementlearning AT tareksayed realtimesafetyoptimizationofconnectedvehicletrajectoriesusingreinforcementlearning |