CityPlanner: A Sandbox Agent for Executable Urban Planning

Researchers introduced CityPlanner, a sandbox-agent framework for executable urban planning, detailed in arXiv paper 2609.09578v1. CityPlanner uses UrbanSandbox, a unified file-based environment where agents inspect task files, generate plans, run evaluators, and revise decisions, plus atomic-task reinforcement learning that splits long trajectories into BuildPlan for initial construction and ImprovePlan for feedback-based refinement. Experiments on a real-world benchmark show CityPlanner consistently outperforms heuristic, task-specific RL, and general LLM-agent baselines, with ablations confirming the contributions of UrbanSandbox, atomic-task RL, and iterative deployment; code and dataset are released at https://anonymous.4open.science/r/co-agent-C1C8.

arXiv:2609.09578v1 Announce Type: new Abstract: Urban planning is a real-world spatial optimization problem that requires selecting feasible actions from large candidate spaces under practical objectives such as cost and service quality. Existing optimization and reinforcement learning methods are effective for fixed formulations, but often depend on task-specific representations and constraint handling. We propose \emph{CityPlanner}, a sandbox-agent framework for executable urban planning. CityPlanner introduces \emph{UrbanSandbox}, a unified file-based environment where agents inspect task files, generate plans, run evaluators, and revise decisions based on executable feedback. To make learning tractable, we further propose atomic-task reinforcement learning, which decomposes long sandbox trajectories into \emph{BuildPlan} for initial construction and \emph{ImprovePlan} for feedback-based refinement. Experiments on a real-world benchmark show that CityPlanner consistently outperforms heuristic, task-specific RL, and general LLM-agent baselines. Ablations verify the contributions of UrbanSandbox, atomic-task RL, and iterative deployment. We release the code and dataset at https://anonymous.4open.science/r/co-agent-C1C8