File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Dispatch of autonomous vehicles for taxi services: A deep reinforcement learning approach

TitleDispatch of autonomous vehicles for taxi services: A deep reinforcement learning approach
Authors
KeywordsAutonomous vehicles
Deep reinforcement learning
Actor-critic algorithm
Demand rebalancing
Ride sharing
Policy-gradient
Taxi dispatching
Issue Date2020
Citation
Transportation Research Part C: Emerging Technologies, 2020, v. 115, article no. 102626 How to Cite?
Abstract© 2020 Elsevier Ltd In this paper, we define and investigate a novel model-free deep reinforcement learning framework to solve the taxi dispatch problem. The framework can be used to redistribute vehicles when the travel demand and taxi supply is either spatially or temporally imbalanced in a transportation network. While previous works mostly focus on using model-based methods, the goal of this paper is to explore the policy-based deep reinforcement learning algorithm as a model-free method to optimize the rebalancing strategy. In particular, we propose an actor-critic algorithm with feed-forward neural networks as approximations of both policy and value functions, where the policy function provides the optimal dispatch strategy and the value function estimates the expected costs at each time stamp. Our numerical studies show that the algorithm converges to the theoretical upper bound with less than 4% optimality gap, whether the system dynamics are deterministic or stochastic. We also investigate the scenario where we consider user priority and fairness, and the results indicate that our learned policy is capable of producing a superior strategy that balances equity, cancellation, and level of service when user priority is considered.
Persistent Identifierhttp://hdl.handle.net/10722/296212
ISSN
2023 Impact Factor: 7.6
2023 SCImago Journal Rankings: 2.860
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorMao, Chao-
dc.contributor.authorLiu, Yulin-
dc.contributor.authorShen, Zuo Jun (Max)-
dc.date.accessioned2021-02-11T04:53:04Z-
dc.date.available2021-02-11T04:53:04Z-
dc.date.issued2020-
dc.identifier.citationTransportation Research Part C: Emerging Technologies, 2020, v. 115, article no. 102626-
dc.identifier.issn0968-090X-
dc.identifier.urihttp://hdl.handle.net/10722/296212-
dc.description.abstract© 2020 Elsevier Ltd In this paper, we define and investigate a novel model-free deep reinforcement learning framework to solve the taxi dispatch problem. The framework can be used to redistribute vehicles when the travel demand and taxi supply is either spatially or temporally imbalanced in a transportation network. While previous works mostly focus on using model-based methods, the goal of this paper is to explore the policy-based deep reinforcement learning algorithm as a model-free method to optimize the rebalancing strategy. In particular, we propose an actor-critic algorithm with feed-forward neural networks as approximations of both policy and value functions, where the policy function provides the optimal dispatch strategy and the value function estimates the expected costs at each time stamp. Our numerical studies show that the algorithm converges to the theoretical upper bound with less than 4% optimality gap, whether the system dynamics are deterministic or stochastic. We also investigate the scenario where we consider user priority and fairness, and the results indicate that our learned policy is capable of producing a superior strategy that balances equity, cancellation, and level of service when user priority is considered.-
dc.languageeng-
dc.relation.ispartofTransportation Research Part C: Emerging Technologies-
dc.subjectAutonomous vehicles-
dc.subjectDeep reinforcement learning-
dc.subjectActor-critic algorithm-
dc.subjectDemand rebalancing-
dc.subjectRide sharing-
dc.subjectPolicy-gradient-
dc.subjectTaxi dispatching-
dc.titleDispatch of autonomous vehicles for taxi services: A deep reinforcement learning approach-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1016/j.trc.2020.102626-
dc.identifier.scopuseid_2-s2.0-85082970900-
dc.identifier.volume115-
dc.identifier.spagearticle no. 102626-
dc.identifier.epagearticle no. 102626-
dc.identifier.isiWOS:000535426900018-
dc.identifier.issnl0968-090X-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats