Building A Transparent Multi-Agent AI System to Help Farmers Make Better Precision-Farming Decisions Using Digital Twins in Changing Climates.
Author -1 Author-2
Saurabh D. Chavhan Aniket R. Sarode
Assistant Professor Assistant Professor
Department of Computer Science Department of Computer Science
Vidya Bharati Mahavidyalaya ,Amravati Vidya Bharati Mahavidyalaya ,Amravati
saurabhchavhan0705@gmail.com aniketsarode.r@gmail.com
ABSTRACT
Husbandry is decreasingly affected by climate query, resource constraints, and rising profitable pressure on growers and agribusiness associations. While artificial intelligence( AI) has been espoused in perfection husbandry to optimize inputs and ameliorate productivity, utmost being systems remain reactive, task-specific, and delicate to interpret, limiting their strategic value and relinquishment. This study proposes the design and evaluation of an resolvable, multi-agent, thing- driven Agentic AI frame that autonomously manages perfection husbandry under climate query using digital binary simulation and mortal- in- the- circle decision control. The frame integrates multiple intelligent agents responsible for soil monitoring, crop health analysis, climate assessment, irrigation planning, and pest operation. A digital twin of the ranch terrain is employed to pretend and validate AI opinions before real- world prosecution, reducing functional threat. resolvable AI mechanisms and mortal oversight are bedded to enhance translucency, trust, and directorial responsibility. Simulation- grounded evaluation under varying climate scripts demonstrates bettered resource effectiveness, yield stability, and decision translucency compared to traditional AI- grounded perfection husbandry systems. From a operation perspective, the proposed frame supports strategic decision- timber, sustainability pretensions, and long- term adaptability in agrarian operations.
Keywords:
Agentic AI, Smart Precision Farming, Multi-Agent Intelligence, Explainable AI (XAI), Digital Twin Technology, Climate Variability, Human-Guided Decision Making, Autonomous Farm Management, Sustainable Agriculture
