A Distributionally Robust Reinforcement Learning Framework for Constrained Urban EV Dispatch
arXiv:2604.25848v2 Announce Type: replace Abstract: We study city-scale control of electric-vehicle (EV) ride-hailing fleets where dispatch, repositioning, and charging decisions must respect charger and feeder limits under uncertain, spatially correlated demand and travel times. We formulate the problem as a hex-grid semi-Markov decision process (semi-MDP) with mixed actions -- discrete actions for serving, repositioning, and charging, together with continuous charging power -- and variable action durations. To guarantee physical feasibility during both training and deployment, the policy lea
The rapid expansion of electric vehicle fleets and ride-hailing services, coupled with increasing computational power, makes sophisticated city-scale optimization both necessary and feasible now.
This work addresses critical infrastructure and operational challenges in urban EV deployment, which is a major component of future smart cities and sustainable transportation.
The ability to manage complex EV fleets under uncertainty with physical constraints through robust reinforcement learning will enable more efficient and reliable urban mobility systems.
- · Ride-hailing companies
- · Smart city developers
- · AI/ML solution providers
- · Electric vehicle manufacturers
- · Traditional taxi services
- · Inefficient urban logistics operators
Improved efficiency and profitability of urban EV fleets due to optimized dispatch, repositioning, and charging strategies.
Reduced urban congestion and emissions as EV ride-hailing becomes more ubiquitous and reliably managed.
Accelerated adoption of EVs in urban environments, necessitating increased grid capacity and smart charging infrastructure.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI