Enhanced Delta-tolling: Traffic Optimization via Policy Gradient Reinforcement Learning Conference Paper uri icon

abstract

  • © 2018 IEEE. In the micro-tolling paradigm, a centralized system manager sets different toll values for each link in a given traffic network with the objective of optimizing the system's performance. A recently proposed micro-tolling scheme, denoted Delta-tolling, was shown to yield up to 32% reduction in total travel time when compared to a no-toll scheme. Delta-tolling, computes a toll value for each link in a given network based on two global parameters: \beta which is a proportional parameter and R which controls the rate of toll change over time. In this paper, we propose to generalize Delta-tolling such that it would consider different R and \beta parameters for each link. a policy gradient reinforcement learning algorithm is used in order to tune this high-dimensional optimization problem. The results show that such a variant of Delta-tolling far surpasses the original Delta-tolling scheme, yielding up to 38% reduced system travel time compared to the original Delta-tolling scheme.

name of conference

  • 2018 21st International Conference on Intelligent Transportation Systems (ITSC)

published proceedings

  • 2018 21ST INTERNATIONAL CONFERENCE ON INTELLIGENT TRANSPORTATION SYSTEMS (ITSC)

author list (cited authors)

  • Mirzaei, H., Sharon, G., Boyles, S., Givargis, T., & Stone, P.

citation count

  • 8

complete list of authors

  • Mirzaei, Hamid||Sharon, Guni||Boyles, Stephen||Givargis, Tony||Stone, Peter

publication date

  • January 2018