Air Quality Monitoring and Forecasting Using Smart Drones and Recurrent Neural Network for Sustainable Development in Chennai City

Air quality monitoring is essential due to increased pollution and health hazards in today's world. Several studies are being conducted to evaluate air quality in metropolitan cities, but certain key research gaps such as the inconsistency in measuring multiple pollutants, and the capability of the models to deal with multiple pollutant concentrations remain unaddressed. Initially, a Smart Drone is designed to monitor 9 different pollutants present in the air in a Metropolitan City. Air quality parameters are monitored using drones in and around solid waste dump yards. Real-time data for training is used for the bidirectional gated recurrent unit (Bi-GRU) network model development. The network model estimates the concentration of different contaminants in the air based on the training, delivering information about air quality in various places. The model yielded average RMSE values of 1.63, 1.35, and 1.47 for solid waste dumpsites, residential areas, and industrial areas. Similarly, the average MAE of the proposed model for waste disposal sites is 0.25, 0.89 for the industrial sector, and 0.75 for the residential area. This combined method of deploying smart drones to monitor air quality and a deep learning model to forecasting it will aid in urban sustainable planning and development.