# PDDLGym: Gym Environments from PDDL Problems

arXiv: https://arxiv.org/abs/2002.06432 · HTML: https://mlsanity.com/paper/2002.06432/

## Key facts

- Cited by 45 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Feb 2020.

## Built on by (most cited first, 30 of 45)

- [AgentBoard: An Analytical Evaluation Board of Multi-turn LLM Agents](https://mlsanity.com/paper/2401.13178/) (2024)
- [TASKOGRAPHY: Evaluating robot task planning over large 3D scene graphs](https://mlsanity.com/paper/2207.05006/) (2022)
- [DELTA: Decomposed Efficient Long-Term Robot Task Planning using Large Language Models](https://mlsanity.com/paper/2404.03275/) (2024)
- [Vision-Language Interpreter for Robot Task Planning](https://mlsanity.com/paper/2311.00967/) (2023)
- [Classical Planning in Deep Latent Space](https://mlsanity.com/paper/2107.00110/) (2021)
- [Reinforcement Learning for Classical Planning: Viewing Heuristics as Dense Reward Generators](https://mlsanity.com/paper/2109.14830/) (2021)
- [Planning in the Dark: LLM-Symbolic Planning Pipeline without Experts](https://mlsanity.com/paper/2409.15915/) (2024)
- [NeSyC: A Neuro-symbolic Continual Learner For Complex Embodied Tasks In Open Domains](https://mlsanity.com/paper/2503.00870/) (2025)
- [Explainable Reinforcement Learning via Model Transforms](https://mlsanity.com/paper/2209.12006/) (2022)
- [Goal Recognition as Reinforcement Learning](https://mlsanity.com/paper/2202.06356/) (2022)
- [PG3: Policy-Guided Planning for Generalized Policy Generation](https://mlsanity.com/paper/2204.10420/) (2022)
- [pyRDDLGym: From RDDL to Gym Environments](https://mlsanity.com/paper/2211.05939/) (2022)
- [MinePlanner: A Benchmark for Long-Horizon Planning in Large Minecraft Worlds](https://mlsanity.com/paper/2312.12891/) (2023)
- [Simulation to Rules: A Dual-VLM Framework for Formal Visual Planning](https://mlsanity.com/paper/2510.03182/) (2025)
- [NeSyPr: Neurosymbolic Proceduralization For Efficient Embodied Reasoning](https://mlsanity.com/paper/2510.19429/) (2025)
- [Open-Ended Goal Inference through Actions and Language for Human-Robot Collaboration](https://mlsanity.com/paper/2512.04453/) (2025)
- [NovelGym: A Flexible Ecosystem for Hybrid Planning and Learning Agents Designed for Open Worlds](https://mlsanity.com/paper/2401.03546/) (2024)
- [Context Matters! Relaxing Goals with LLMs for Feasible 3D Scene Planning](https://mlsanity.com/paper/2506.15828/) (2025)
- [Learning Lifted Action Models from Unsupervised Visual Traces](https://mlsanity.com/paper/2604.19043/) (2026)
- [Learning First-Order Symbolic Planning Representations That Are Grounded](https://mlsanity.com/paper/2204.11902/) (2022)
- [An Extensive Evaluation of PDDL Capabilities in off-the-shelf LLMs](https://mlsanity.com/paper/2502.20175/) (2025)
- [Under the Influence: Quantifying Persuasion and Vigilance in Large Language Models](https://mlsanity.com/paper/2602.21262/) (2026)
- [ConMem: Structured Memory-Guided Adaptation in Training-Free Multi-Agent Systems](https://mlsanity.com/paper/2606.08702/) (2026)
- [Learning Causal Models of Autonomous Agents using Interventions](https://mlsanity.com/paper/2108.09586/) (2021)
- [Learning Hierarchical Domain Models Through Environment-Grounded Interaction](https://mlsanity.com/paper/2505.13497/) (2025)
- [HDDLGym: A Tool for Studying Multi-Agent Hierarchical Problems Defined in HDDL with OpenAI Gym](https://mlsanity.com/paper/2505.22597/) (2025)
- [Diverse Planning with Simulators via Linear Temporal Logic](https://mlsanity.com/paper/2510.17418/) (2025)
- [Robots Influencing Humans to Reveal their Goals during Collaboration and Competition](https://mlsanity.com/paper/2609.05519/) (2026)
- [Neural NID Rules](https://mlsanity.com/paper/2202.06036/) (2022)
- [The Sandbox Environment for Generalizable Agent Research (SEGAR)](https://mlsanity.com/paper/2203.10351/) (2022)

## Abstract (arXiv)

We present PDDLGym, a framework that automatically constructs OpenAI Gym environments from PDDL domains and problems. Observations and actions in PDDLGym are relational, making the framework particularly well-suited for research in relational reinforcement learning and relational sequential decision-making. PDDLGym is also useful as a generic framework for rapidly building numerous, diverse benchmarks from a concise and familiar specification language. We discuss design decisions and implementation details, and also illustrate empirical variations between the 20 built-in environments in terms of planning and model-learning difficulty. We hope that PDDLGym will facilitate bridge-building between the reinforcement learning community (from which Gym emerged) and the AI planning community (which produced PDDL). We look forward to gathering feedback from all those interested and expanding the set of available environments and features accordingly. Code: https://github.com/tomsilver/pddlgym