Enhancing decision support tools in agroforestry: Integrating farmers’ perspectives on goals, challenges, and experience

Engineering notes

Key topics: autonomous driving. See the paper for implementation details and experimental results.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为端到端自动驾驶、BEV感知、3D目标检测、轨迹预测、路径规划、LiDAR感知等高价值论文补充中文说明。

Original abstract

Introduction Agroforestry systems offer substantial environmental, economic, and social benefits, yet their adoption is constrained by a lack of decision support tools aligned with farmer workflows. This study aimed to characterize farmer typologies, assess prioritization gaps between practitioners and researchers, and develop an Artificial Intelligence (AI)-guided design framework operationalizing farmer-derived logic. Methods We conducted semi-structured surveys with 23 horticulturists and 29 researchers, mapped 859 existing models against farmer requirements, and deployed a multi-layered AI prompt architecture. Results Results revealed four distinct farmer typologies (Soil fertility improvers, Large-scale integrators, Nature protectors, Produce diversifiers) with diverse goals often overestimated by researchers. Existing models showed critical gaps, supporting only 41% of farmer-derived conditional statements, with 64% of required input data inaccessible to practitioners. To address this, we developed and published an AI-guided decision support framework that translates farmer narrative logic into verifiable design workflows. Discussion We conclude that bridging the evidence-practice gap requires reorienting tool development from discipline-centric modeling toward workflow-centric design that integrates local knowledge. Hybrid approaches combining mechanistic constraints with AI-enabled adaptation offer a promising pathway to democratize expert agroforestry knowledge while ensuring epistemic transparency through robust verification architectures.

Links and sources

• PDF from original source

Comments