Scholarlink

<p>Accurate livestock localization in smart farming is challenging due to suboptimal anchor node utilization, collinearity problem and poor UAV-based path planning, leading to reduced localization accuracy and increased energy consumption. Balancing localization accuracy, energy efficiency, and full coverage in large-scale, resource-constrained deployments remains unresolved. To overcome these issues, we propose a novel livestock localization and path-planning framework (LivLocPath) that integrates collaborative adaptive weighted trilateration (CAWT), hybrid generalized learning perturbation equilibrium optimization (GLPEO), and modified human memory optimization (MHMO) algorithms for precise livestock monitoring, optimized anchor node deployment, and energy-efficient UAV-based path planning. This methodology introduces several key innovations: (1) dynamic adaptive weighting in trilateration to prioritize anchor nodes based on real-time signal strength, reliability, and environmental factors; (2) GLPEO-based anchor node optimization, which combines adaptive perturbation and equilibrium strategies to minimize localization errors and avoid local minima; and (3) memory-driven MHMO for refining anchor node deployment and UAV path planning using historical configurations to accelerate convergence and reduce redundant calculations. Furthermore, we introduce dynamic clustering-based UAV path planning, prioritizing high-activity clusters and optimizing energy consumption while ensuring complete area coverage. The proposed CAWT–GLPEO–MHMO LivLocPath framework was evaluated through extensive simulations under varying anchor node densities. The system achieved path efficiency values of 0.86, 0.88, 0.86, 0.87, and 0.85, corresponding to five different anchor node densities, demonstrating consistently near-optimal UAV trajectory performance. Furthermore, the framework attained a coverage efficiency of 98% and an average localization error of 0.7&#xa0;m, confirming its capability to deliver accurate and energy-efficient livestock localization and monitoring in smart farming environments.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Hybrid Optimization Framework for Livestock Localization in Smart Farms with Collaborative Trilateration and Memory-Assisted Path Planning

  • 期刊论文

  • 发表日期：

    2026年1月10日

查看全文
  • S. P. Subotha,
  • M. Shunmugathammal,
  • K. Shanthi,
  • L. Femila

摘要

Accurate livestock localization in smart farming is challenging due to suboptimal anchor node utilization, collinearity problem and poor UAV-based path planning, leading to reduced localization accuracy and increased energy consumption. Balancing localization accuracy, energy efficiency, and full coverage in large-scale, resource-constrained deployments remains unresolved. To overcome these issues, we propose a novel livestock localization and path-planning framework (LivLocPath) that integrates collaborative adaptive weighted trilateration (CAWT), hybrid generalized learning perturbation equilibrium optimization (GLPEO), and modified human memory optimization (MHMO) algorithms for precise livestock monitoring, optimized anchor node deployment, and energy-efficient UAV-based path planning. This methodology introduces several key innovations: (1) dynamic adaptive weighting in trilateration to prioritize anchor nodes based on real-time signal strength, reliability, and environmental factors; (2) GLPEO-based anchor node optimization, which combines adaptive perturbation and equilibrium strategies to minimize localization errors and avoid local minima; and (3) memory-driven MHMO for refining anchor node deployment and UAV path planning using historical configurations to accelerate convergence and reduce redundant calculations. Furthermore, we introduce dynamic clustering-based UAV path planning, prioritizing high-activity clusters and optimizing energy consumption while ensuring complete area coverage. The proposed CAWT–GLPEO–MHMO LivLocPath framework was evaluated through extensive simulations under varying anchor node densities. The system achieved path efficiency values of 0.86, 0.88, 0.86, 0.87, and 0.85, corresponding to five different anchor node densities, demonstrating consistently near-optimal UAV trajectory performance. Furthermore, the framework attained a coverage efficiency of 98% and an average localization error of 0.7 m, confirming its capability to deliver accurate and energy-efficient livestock localization and monitoring in smart farming environments.