Citation

Sungheetha, Akey and R., Rajesh Sharma and Mahapatra, Sheila (2026) Neuromorphic-Enhanced Smart Agricultural Intelligence: An Explainable AI Framework for Sustainable IoT-Driven Crop Yield Optimization with Green Energy Integration. Procedia Computer Science, 283. pp. 5487-5496. ISSN 18770509

Text 15.pdf - Published Version Restricted to Repository staff only Download (405kB)

Abstract

This paper presents the Neuromorphic Explainable Sensor-Fused Artificial Intelligence (NESAI) framework, a theoretically constructed computational model integrating spike-based neural dynamics, multi-modal sensor fusion, and layer-wise relevance propagation for precision agriculture. The framework is formulated entirely within theoretical science computing paradigms, wherein all numerical outcomes are derived through mathematical simulation of biological plausibility in spiking neural networks (SNNs) governed by spike-timing-dependent plasticity (STDP) learning rules with potentiation rate A+ = 0.01 and depression rate A− = 0.012. Sensor modalities encompassing soil volumetric water content (12.3–89.7% VWC), nitrogen concentration (12.4–287.6 ppm), photosynthetically active radiation (0–2347 µmol/m2/s), and atmospheric parameters are mathematically fused under optimised weight vector β = [0.347, 0.238, 0.184, 0.231] across a synthetic dataset of 47,832 computational samples spanning an 18-month agronomic cycle. Theoretical evaluation demonstrates yield prediction accuracy of 92.6%, event-driven system power of 5.3 W, decision latency of 47 ms, and water conservation efficiency of 76.8%. STDP synaptic weight convergence to 0.547±0.083 at 15,000 training samples confirms stable temporal feature selectivity. The explainability module assigns 34.7% relevance to soil moisture, enabling simulated farmer adoption improvement from 58.6% to 91.2%, demonstrating the theoretical viability of neuromorphicXAI integration for sustainable agriculture computing.

Item Type:	Article
Uncontrolled Keywords:	Internet of Things, Neuromorphic computing, Sustainable agriculture
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	04 Sep 2026 07:05
Last Modified:	04 Sep 2026 07:05
URII:	http://shdl.mmu.edu.my/id/eprint/16733

Downloads

Edit (login required)