Scholarlink

<p>The persistent challenge of radiation fog at Amritsar Airport, exacerbated by smog from agricultural and urban pollutants, necessitates innovative solutions to enhance runway visibility. This study proposes an AI-optimized UAV swarm system integrated with UV-C radiation to disperse pollutant-laden fog and improve the runway visual range (RVR). By coupling MATLAB/Simulink (for simulating 6-DOF UAV dynamics) with ANSYS Fluent (for modeling fog microphysics), the framework employs deep reinforcement learning (DRL) to dynamically optimize swarm paths based on real-time LiDAR and RVR data. Results demonstrate that a 4-UAV swarm clears the fog in the Touchdown Zone (TDZ) within 5.06&#xa0;min, achieving an 80% reduction in required visual range (RVR) improvement (from 450 to 810&#xa0;m) under UV energy density of 0.5&#xa0;kW/m<sup>2</sup>, validated against Amritsar Airport data with a 3.1% error margin. The system reduces operational costs by 95% (from 800 per sortie) and CO<sub>2</sub> emissions by 62% compared to traditional methods, aligning with SDGs 9 and 13. This work bridges the fields of UAV aerodynamics, AI, and environmental science, offering a scalable and sustainable model for global airports to combat low-visibility crises.</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

AI-Optimized UAV Swarm System for Radiation Fog Dispersal and Runway Visibility Enhancement at Airport

  • 期刊论文

  • 发表日期：

    2025年12月18日

查看全文
  • Saifullah Khalid,
  • Shashi Kant Gupta,
  • Midhun Chakkaravarthy,
  • Dharmendra Prakash,
  • Alkesh Agrawal,
  • D. K. Nishad

摘要

The persistent challenge of radiation fog at Amritsar Airport, exacerbated by smog from agricultural and urban pollutants, necessitates innovative solutions to enhance runway visibility. This study proposes an AI-optimized UAV swarm system integrated with UV-C radiation to disperse pollutant-laden fog and improve the runway visual range (RVR). By coupling MATLAB/Simulink (for simulating 6-DOF UAV dynamics) with ANSYS Fluent (for modeling fog microphysics), the framework employs deep reinforcement learning (DRL) to dynamically optimize swarm paths based on real-time LiDAR and RVR data. Results demonstrate that a 4-UAV swarm clears the fog in the Touchdown Zone (TDZ) within 5.06 min, achieving an 80% reduction in required visual range (RVR) improvement (from 450 to 810 m) under UV energy density of 0.5 kW/m2, validated against Amritsar Airport data with a 3.1% error margin. The system reduces operational costs by 95% (from 800 per sortie) and CO2 emissions by 62% compared to traditional methods, aligning with SDGs 9 and 13. This work bridges the fields of UAV aerodynamics, AI, and environmental science, offering a scalable and sustainable model for global airports to combat low-visibility crises.