The New Agronomists: Language Models are Experts in Crop Management

TL;DR

This paper presents a novel crop management system that integrates reinforcement learning with language models and crop simulations, achieving superior performance and economic benefits in maize cultivation in different regions.

Contribution

It introduces a unique combination of deep RL, language models, and crop simulations, enhancing decision-making in crop management beyond prior neural network-based approaches.

Findings

Over 49% increase in economic profit

Superior learning capabilities of the language model

Reduced environmental impact compared to baselines

Abstract

Crop management plays a crucial role in determining crop yield, economic profitability, and environmental sustainability. Despite the availability of management guidelines, optimizing these practices remains a complex and multifaceted challenge. In response, previous studies have explored using reinforcement learning with crop simulators, typically employing simple neural-network-based reinforcement learning (RL) agents. Building on this foundation, this paper introduces a more advanced intelligent crop management system. This system uniquely combines RL, a language model (LM), and crop simulations facilitated by the Decision Support System for Agrotechnology Transfer (DSSAT). We utilize deep RL, specifically a deep Q-network, to train management policies that process numerous state variables from the simulator as observations. A novel aspect of our approach is the conversion of these state variables into more informative language, facilitating the language model's capacity to understand states and explore optimal management practices. The empirical results reveal that the LM exhibits superior learning capabilities. Through simulation experiments with maize crops in Florida (US) and Zaragoza (Spain), the LM not only achieves state-of-the-art performance under various evaluation metrics but also demonstrates a remarkable improvement of over 49\% in economic profit, coupled with reduced environmental impact when compared to baseline methods. Our code is available at \url{https://github.com/jingwu6/LM_AG}.