CityLearn provides a modular simulation platform for energy management research using reinforcement learning. Users can define multi-zone building clusters, configure HVAC systems, storage units, and renewable sources, then train RL agents against demand response events. The environment exposes state observations like temperatures, load profiles, and energy prices, while actions control setpoints and storage dispatch. A flexible reward API allows custom metrics—such as cost savings or emission reductions—and logging utilities support performance analysis. CityLearn is ideal for benchmarking, curriculum learning, and developing novel control strategies in a reproducible research framework.
CityLearn Core Features
Configurable multi-zone building and microgrid simulation
Demand response event modeling
Customizable reward function API
Baseline agent implementations
Detailed logging and analytics tools
Scenario and dataset management
CityLearn Pro & Cons
The Cons
Primarily focused on training and simulation, may require integration with actual robotic hardware for deployment.
Relies on availability of high-quality datasets for training realistic navigation policies.
No pricing or commercial support information available.
The Pros
Enables training across large, city-sized, real-world environments with extreme environmental changes.
Utilizes compact bimodal image representations for sample-efficient learning, reducing training time significantly compared to raw image methods.
Supports generalization across day/night and seasonal transitions, improving robustness of navigation policies.
Open source with publicly available code and datasets.
