Analysis and Control of Autonomous Mobility-on-Demand Systems

Zardini, Gioele; Lanzetti, Nicolas; Pavone, Marco; Frazzoli, Emilio

doi:10.48550/arxiv.2106.14827

Cited by 4 publications

(9 citation statements)

References 127 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the advent of autonomous vehicles, vehicle rebalancing problems have been studied extensively for AMoD systems as well in recent years [5]. A fluid model was utilized to model passengers and vehicles, and an optimal rebalancing policy was developed by solving a linear program [22].…”

Section: A Vehicle Rebalancingmentioning

confidence: 99%

See 1 more Smart Citation

Data-Driven Vehicle Rebalancing With Predictive Prescriptions in the Ride-Hailing System

Guo

Wang

Zhao

2022

IEEE Open J. Intell. Transp. Syst.

View full text Add to dashboard Cite

Rebalancing vacant vehicles is one of the most critical strategies in ride-hailing operations. An effective rebalancing strategy can significantly reduce empty miles traveled and reduce customer wait times by better matching supply and demand. While the supply (vehicles) is usually known to the system, future passenger demand is uncertain. There are two ways to handle uncertainty. First, the point-prediction-driven optimization framework involves predicting the future demand and then producing rebalancing decisions based on the predicted demand. Second, the data-driven optimization approaches directly prescribe rebalancing decisions from data. In this study, a predictive prescription framework is introduced to this problem, where the benefits of predictive and data-driven optimization models are combined. Based on a state-of-the-art vehicle rebalancing model, the matching-integrated vehicle rebalancing (MIVR) model, predictive prescriptions are introduced to handle demand uncertainty. Model performances are evaluated using real-world simulations with New York City (NYC) ride-hailing data under four demand scenarios. When demand can be accurately predicted, a point-prediction-driven optimization framework should be adapted. The proposed predictive prescription models achieve shorter customer wait times over the point-prediction-driven optimization models when future demand predictions are not so accurate, and achieve a competitive performance with respect to the cutting-edge robust optimization models.

show abstract

Section: A Vehicle Rebalancingmentioning

confidence: 99%

“…With technological advances in the field of autonomous driving in the past decade, Autonomous Mobility-on-Demand (AMoD) systems are becoming a reality. With a fleet of autonomous vehicles (AVs), centralized control and planning of vehicles become more vital to efficient operations [5].…”

Section: Introductionmentioning

confidence: 99%

Data-Driven Vehicle Rebalancing With Predictive Prescriptions in the Ride-Hailing System

Guo

Wang

Zhao

2022

IEEE Open J. Intell. Transp. Syst.

View full text Add to dashboard Cite

show abstract

“…Autonomous mobility-on-demand (AMoD) system is one of the most promising energy-efficient transportation solutions as it provides people one-way rides from their origins to destinations [1]- [3]. Electric vehicles (EVs) are being adopted worldwide for environmental and economical benefits [4], and AMoD systems embrace this trend without exception.…”

Section: Introductionmentioning

confidence: 99%

“…Electric vehicles (EVs) are being adopted worldwide for environmental and economical benefits [4], and AMoD systems embrace this trend without exception. However, the trips sporadically appear, and the origins and destinations are asymmetrically distributed and hard to predict in AMoD systems [3], [5]. Such spatial-temporal nature of urban mobility increases the management difficulty of a large-scale vehicle fleet and makes the system sensitive to disturbances [5]- [7].…”

Section: Introductionmentioning

confidence: 99%

A Robust and Constrained Multi-Agent Reinforcement Learning Framework for Electric Vehicle AMoD Systems

He¹,

Wang²,

Han³

et al. 2022

Preprint

View full text Add to dashboard Cite

Electric vehicles (EVs) play critical roles in autonomous mobility-on-demand (AMoD) systems, but their unique charging patterns increase the model uncertainties in AMoD systems (e.g. state transition probability). Since there usually exists a mismatch between the training and test (true) environments, incorporating model uncertainty into system design is of critical importance in real-world applications. However, model uncertainties have not been considered explicitly in EV AMoD system rebalancing by existing literature yet and remain an urgent and challenging task. In this work, we design a robust and constrained multi-agent reinforcement learning (MARL) framework with transition kernel uncertainty for the EV rebalancing and charging problem. We then propose a robust and constrained MARL algorithm (ROCOMA) that trains a robust EV rebalancing policy to balance the supply-demand ratio and the charging utilization rate across the whole city under state transition uncertainty. Experiments show that the ROCOMA can learn an effective and robust rebalancing policy. It outperforms non-robust MARL methods when there are model uncertainties. It increases the system fairness by 19.6% and decreases the rebalancing costs by 75.8%.

show abstract

“…The approach is based on the microscopic level, which uses a fluid-approximated model. A more comprehensive review of the research efforts on AMoD is provided in [11]. There are also several review papers providing a good summary of shared mobility, e.g., see [12]- [18].…”

Section: Introductionmentioning

confidence: 99%

Congestion-Aware Routing, Rebalancing, and Charging Scheduling for Electric Autonomous Mobility-on-Demand System

Bang¹,

Malikopoulos²

2021

Preprint

View full text Add to dashboard Cite

In this paper, we investigate the problem of routing, rebalancing, and charging for autonomous electric mobility-on-demand systems with respect to traffic congestion. We analyze the problem at the macroscopical level and use a volume-delay function to capture traffic congestion. To solve this problem, we first formulate an optimization problem for routing and rebalancing. Then, we present heuristic algorithms to find the loop of the flow and examine the energy constraints within the resulting loop. We impose charging constraints on re-routing problem so that the new solution satisfies the energy constraint. Finally, we verify the effectiveness of our method through simulation.

show abstract

Analysis and Control of Autonomous Mobility-on-Demand Systems

Cited by 4 publications

References 127 publications

Data-Driven Vehicle Rebalancing With Predictive Prescriptions in the Ride-Hailing System

Data-Driven Vehicle Rebalancing With Predictive Prescriptions in the Ride-Hailing System

A Robust and Constrained Multi-Agent Reinforcement Learning Framework for Electric Vehicle AMoD Systems

Congestion-Aware Routing, Rebalancing, and Charging Scheduling for Electric Autonomous Mobility-on-Demand System

Contact Info

Product

Resources

About