# Real-time Control of Electric Autonomous Mobility-on-Demand Systems via Graph Reinforcement Learning

arXiv: https://arxiv.org/abs/2311.05780 · HTML: https://mlsanity.com/paper/2311.05780/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 23 references that are in the map.
- First posted on arXiv 9 Nov 2023.

## Builds on (most cited first, 15 of 15)

- [Semi-Supervised Classification with Graph Convolutional Networks](https://mlsanity.com/paper/1609.02907/) (2016)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Deep Reinforcement Learning for Multi-Driver Vehicle Dispatching and Repositioning Problem](https://mlsanity.com/paper/1911.11260/) (2019)
- [Graph Neural Network Reinforcement Learning for Autonomous Mobility-on-Demand Systems](https://mlsanity.com/paper/2104.11434/) (2021)
- [Data-Driven Model Predictive Control of Autonomous Mobility-on-Demand Systems](https://mlsanity.com/paper/1709.07032/) (2017)
- [Graph Meta-Reinforcement Learning for Transferable Autonomous Mobility-on-Demand](https://mlsanity.com/paper/2202.07147/) (2022)
- [Graph Reinforcement Learning for Network Control via Bi-Level Optimization](https://mlsanity.com/paper/2305.09129/) (2023)
- [A Reinforcement Learning Approach for Rebalancing Electric Vehicle Sharing Systems](https://mlsanity.com/paper/2010.02369/) (2020)
- [Analysis and Control of Autonomous Mobility-on-Demand Systems](https://mlsanity.com/paper/2106.14827/) (2021)
- [On the interaction between Autonomous Mobility-on-Demand systems and the power network: models and coordination algorithms](https://mlsanity.com/paper/1709.04906/) (2017)
- [Joint Optimization of Autonomous Electric Vehicle Fleet Operations and Charging Station Siting](https://mlsanity.com/paper/2107.00165/) (2021)
- [Stochastic Model Predictive Control for Autonomous Mobility on Demand](https://mlsanity.com/paper/1804.11074/) (2018)
- [Congestion-Aware Routing, Rebalancing, and Charging Scheduling for Electric Autonomous Mobility-on-Demand System](https://mlsanity.com/paper/2110.04911/) (2021)
- [On the Interaction between Autonomous Mobility on Demand Systems and Power Distribution Networks -- An Optimal Power Flow Approach](https://mlsanity.com/paper/1905.00200/) (2019)
- [Dynamic Ridesharing in Peak Travel Periods](https://mlsanity.com/paper/2004.02570/) (2020)

## Built on by (most cited first, 2 of 2)

- [Offline Hierarchical Reinforcement Learning via Inverse Optimization](https://mlsanity.com/paper/2410.07933/) (2024)
- [Robo-taxi Fleet Coordination at Scale via Reinforcement Learning](https://mlsanity.com/paper/2504.06125/) (2025)

## Abstract (arXiv)

Operators of Electric Autonomous Mobility-on-Demand (E-AMoD) fleets need to make several real-time decisions such as matching available vehicles to ride requests, rebalancing idle vehicles to areas of high demand, and charging vehicles to ensure sufficient range. While this problem can be posed as a linear program that optimizes flows over a space-charge-time graph, the size of the resulting optimization problem does not allow for real-time implementation in realistic settings. In this work, we present the E-AMoD control problem through the lens of reinforcement learning and propose a graph network-based framework to achieve drastically improved scalability and superior performance over heuristics. Specifically, we adopt a bi-level formulation where we (1) leverage a graph network-based RL agent to specify a desired next state in the space-charge graph, and (2) solve more tractable linear programs to best achieve the desired state while ensuring feasibility. Experiments using real-world data from San Francisco and New York City show that our approach achieves up to 89% of the profits of the theoretically-optimal solution while achieving more than a 100x speedup in computational time. We further highlight promising zero-shot transfer capabilities of our learned policy on tasks such as inter-city generalization and service area expansion, thus showing the utility, scalability, and flexibility of our framework. Finally, our approach outperforms the best domain-specific heuristics with comparable runtimes, with an increase in profits by up to 3.2x.