Genetic Algorithm for Optimizing Meteorological Sensors Placement Using Airborne LiDAR Data

This paper presents a novel method for optimizing meteorological sensor placement over large geographic areas, leveraging genetic algorithm (GA) and classified Light Detection And Ranging (LiDAR) data. The airborne LiDAR data provides detailed surface information necessary for finding the optimal locations for sensor placement. By considering semantic segmentation of surface, relative Sky View Factor (SVF), and surface slope, the method significantly reduces the search space to find suitable locations for sensors placement. The considered GA then uniformly and optimally positions a user-specified number of sensors, guided by optimization constraints within the fitness function. This approach enhances dense meteorological data collection efficiency, ensuring better data quality and coverage in areas such as weather forecasting, climate studies, and pollution monitoring. It reduces redundancy and deployment costs while maximizing sensors exposure to environmental physical processes such as solar radiation and wind.