A Modular and Transferable Reinforcement Learning Framework for the Fleet Rebalancing Problem

Mobility on demand (MoD) systems show great promise in realizing flexible and efficient urban transportation. However, significant technical challenges arise from operational decision making associated with MoD vehicle dispatch and fleet rebalancing. For this reason, operators tend to employ simplified algorithms that have been demonstrated to work well in a particular setting. To help bridge the gap between novel and existing methods, we propose a modular framework for fleet rebalancing based on model-free reinforcement learning (RL) that can leverage an existing dispatch method to minimize system cost. In particular, by treating dispatch as part of the environment dynamics, a centralized agent can learn to intermittently direct the dispatcher to reposition free vehicles and mitigate against fleet imbalance. We formulate RL state and action spaces as distributions over a grid partitioning of the operating area, making the framework scalable and avoiding the complexities associated with multiagent RL. Numerical experiments, using real-world trip and network data, demonstrate that this approach has several distinct advantages over baseline methods including: improved system cost; high degree of adaptability to the selected dispatch method; and the ability to perform scale-invariant transfer learning between problem instances with similar vehicle and request distributions.

[1]  Forrest N. Iandola,et al.  SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1MB model size , 2016, ArXiv.

[2]  Matthew J. Roorda,et al.  Vehicle relocation and staff rebalancing in one-way carsharing systems , 2015 .

[3]  Qing Xiao,et al.  Distributed Fusion-Based Policy Search for Fast Robot Locomotion Learning , 2019, IEEE Computational Intelligence Magazine.

[4]  Klaus Bogenberger,et al.  Comparing Future Autonomous Electric Taxis With an Existing Free-Floating Carsharing System , 2019, IEEE Transactions on Intelligent Transportation Systems.

[5]  Yong Gao,et al.  Optimize taxi driving strategies based on reinforcement learning , 2018, Int. J. Geogr. Inf. Sci..

[6]  Marco Pavone,et al.  Analysis, Control, and Evaluation of Mobility-on-Demand Systems: A Queueing-Theoretical Approach , 2019, IEEE Transactions on Control of Network Systems.

[7]  Mustafa Ammous,et al.  Optimal Cloud-Based Routing With In-Route Charging of Mobility-on-Demand Electric Vehicles , 2019, IEEE Transactions on Intelligent Transportation Systems.

[8]  Kara M. Kockelman,et al.  Dynamic Ride-Sharing and Optimal Fleet Sizing for a System of Shared Autonomous Vehicles , 2015 .

[9]  Dimitri Bertsekas,et al.  Multiagent Reinforcement Learning: Rollout and Policy Iteration , 2021, IEEE/CAA Journal of Automatica Sinica.

[10]  Zhihao Chen,et al.  Optimizing the Profitability and Quality of Service in Carshare Systems Under Demand Uncertainty , 2018, Manuf. Serv. Oper. Manag..

[11]  Christoph Lauer,et al.  REBALANCING AND FLEET SIZING OF MOBILITY-ON-DEMAND NETWORKS WITH COMBINED SIMULATION, OPTIMIZATION AND QUEUEING NETWORK ANALYSIS , 2018, 2018 Winter Simulation Conference (WSC).

[12]  Ge Guo,et al.  A Deep Reinforcement Learning Approach to Ride-Sharing Vehicle Dispatching in Autonomous Mobility-on-Demand Systems , 2022, IEEE Intelligent Transportation Systems Magazine.

[13]  Zhe Xu,et al.  Efficient Large-Scale Fleet Management via Multi-Agent Deep Reinforcement Learning , 2018, KDD.

[14]  Mustafa Ammous,et al.  Fog-Based Multi-Class Dispatching and Charging for Autonomous Electric Mobility On-Demand , 2020, IEEE Transactions on Intelligent Transportation Systems.

[15]  Robert C. Hampshire,et al.  Inventory rebalancing and vehicle routing in bike sharing systems , 2017, Eur. J. Oper. Res..

[16]  Joseph Warrington,et al.  Two-stage stochastic approximation for dynamic rebalancing of shared mobility systems , 2018, Transportation Research Part C: Emerging Technologies.

[17]  Michael W. Levin,et al.  Dynamic User Equilibrium of Mobility-on-Demand System with Linear Programming Rebalancing Strategy , 2019 .

[18]  Frédéric Meunier,et al.  Bike sharing systems: Solving the static rebalancing problem , 2013, Discret. Optim..

[19]  Edsger W. Dijkstra,et al.  A note on two problems in connexion with graphs , 1959, Numerische Mathematik.

[20]  Emilio Frazzoli,et al.  Scalable Model Predictive Control for Autonomous Mobility-on-Demand Systems , 2019, IEEE Transactions on Control Systems Technology.

[21]  Patrick Jaillet,et al.  Rebalancing shared mobility-on-demand systems: A reinforcement learning approach , 2017, 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC).

[22]  Maxime Guériau,et al.  SAMoD: Shared Autonomous Mobility-on-Demand using Decentralized Reinforcement Learning , 2018, 2018 21st International Conference on Intelligent Transportation Systems (ITSC).

[23]  Satish V. Ukkusuri,et al.  Efficient proactive vehicle relocation for on-demand mobility service with recurrent neural networks , 2020 .

[24]  Emilio Frazzoli,et al.  Robotic load balancing for mobility-on-demand systems , 2012, Int. J. Robotics Res..

[25]  Moshe Ben-Akiva,et al.  Assessing the impacts of automated mobility-on-demand through agent-based simulation: A study of Singapore , 2020, Transportation Research Part A: Policy and Practice.

[26]  Vaneet Aggarwal,et al.  DeepPool: Distributed Model-Free Algorithm for Ride-Sharing Using Deep Reinforcement Learning , 2019, IEEE Transactions on Intelligent Transportation Systems.

[27]  Yulin Liu,et al.  Dispatch of autonomous vehicles for taxi services: A deep reinforcement learning approach , 2020, Transportation Research Part C: Emerging Technologies.

[28]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[29]  Alec Radford,et al.  Proximal Policy Optimization Algorithms , 2017, ArXiv.

[30]  Marco Pavone,et al.  Routing autonomous vehicles in congested transportation networks: structural properties and coordination algorithms , 2016, Autonomous Robots.

[31]  Stefan Illgen,et al.  Literature review of the vehicle relocation problem in one-way car sharing networks , 2019, Transportation Research Part B: Methodological.

[32]  Federico Chiariotti,et al.  A Dynamic Approach to Rebalancing Bike-Sharing Systems , 2018, Sensors.

[33]  Marco Pavone,et al.  Control of robotic mobility-on-demand systems: A queueing-theoretical perspective , 2014, Int. J. Robotics Res..

[34]  Ramtin Pedarsani,et al.  Dynamic pricing and fleet management for electric autonomous mobility on demand systems , 2020 .