WOFOSTGym: A Crop Simulator for Learning Annual and Perennial Crop Management Strategies

2025-02-26

AI Summary

This work addresses the challenge of coordinated annual and perennial crop management in single- and multi-farm settings. We propose the first reinforcement learning (RL) agricultural simulation environment supporting 23 annual and 2 perennial crops across farms and multiple years. Unlike existing simulators—limited to annual crops and incapable of supporting multi-farm RL training—our framework unifies realistic agricultural constraints, including partial observability, non-Markovian dynamics, and delayed feedback. Built upon an extended WOFOST mechanistic crop growth model and integrated with standard OpenAI Gym interfaces, it is compatible with mainstream RL algorithms. Experiments demonstrate strong policy generalization across diverse crop varieties and soil types. The environment enables joint optimization of yield, economic return, and environmental sustainability, significantly improving decision robustness and long-term agroecological performance.

Technology Category

• Machine Learning: Reinforcement Learning
• Multiagent Systems: Multiagent Learning
• Search and Optimization: Learning to Search

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Search and Retrieval-Augmented AI: Agentic search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

We introduce WOFOSTGym, a novel crop simulation environment designed to train reinforcement learning (RL) agents to optimize agromanagement decisions for annual and perennial crops in single and multi-farm settings. Effective crop management requires optimizing yield and economic returns while minimizing environmental impact, a complex sequential decision-making problem well suited for RL. However, the lack of simulators for perennial crops in multi-farm contexts has hindered RL applications in this domain. Existing crop simulators also do not support multiple annual crops. WOFOSTGym addresses these gaps by supporting 23 annual crops and two perennial crops, enabling RL agents to learn diverse agromanagement strategies in multi-year, multi-crop, and multi-farm settings. Our simulator offers a suite of challenging tasks for learning under partial observability, non-Markovian dynamics, and delayed feedback. WOFOSTGym's standard RL interface allows researchers without agricultural expertise to explore a wide range of agromanagement problems. Our experiments demonstrate the learned behaviors across various crop varieties and soil types, highlighting WOFOSTGym's potential for advancing RL-driven decision support in agriculture.

Problem

• Optimizing crop management for yield
• Training RL agents for agriculture
• Supporting multi-crop and multi-farm simulations

Innovation

• Reinforcement learning for crop management
• Supports 23 annual crops, 2 perennials
• Multi-farm, multi-crop simulation environment