Sign in Subscribe
Machine learning

Artificial Intelligence Advances Air Quality Monitoring with Embedded Machine Learning

Strategic Insights

— 1 min read

Research from the Department of Science and Technology has highlighted the growing need for accurate and real-time air quality monitoring in urban environments. The study, which utilized a novel embedded machine learning approach with edge computing, demonstrated the effectiveness of low-cost sensor enhancement in improving air quality monitoring and control. The proposed system, which employs the one-rank cuckoo search-driven adaptive support vector machine (ORCS-ASVM), was found to significantly outperform traditional prediction algorithms in terms of error reduction, with detection accuracies of 94.2%, 95.2%, and 94.8% for PM2.5, SO2, and NO2, respectively.

Key Takeaways:

  • The Department of Science and Technology's research emphasizes the need for operational real-time monitoring and management of air pollution in urban environments.
  • Low-cost sensors, while less accurate, are often too costly for mass deployment in IoT environments.
  • The proposed ORCS-ASVM machine learning model enhances the accuracy and reliability of low-cost sensor readings, reducing the amount of computing power required.
  • The system utilizes a Raspberry Pi as the embedded edge device for processing sensor information, allowing for real-time monitoring of multiple air pollutants.
  • Smoothing and normalization of sensor data are common preprocessing methods used to correct sensor errors.
  • The study demonstrates the effectiveness of an embedded machine learning system approach for air quality monitoring and in-situ pollution mitigation.

Statistics:

  • Air pollutants such as PM2.5, SO2, and NO2 achieved detection accuracies of 94.2%, 95.2%, and 94.8%, respectively.
  • The proposed ORCS-ASVM model outperformed traditional prediction algorithms in terms of error reduction.
  • The system utilizes a Raspberry Pi as the embedded edge device for processing sensor information.

Sources:

  • An embedded machine learning system method for air pollution monitoring and control. AIMS Environmental Science, 2025,12(4):576-593.
  • NewsRx. Research from Department of Science and Technology Provides New Study Findings on Machine Learning (An embedded machine learning system method for air pollution monitoring and control). Ecology, Environment & Conservation. September 12, 2025; p 692.

Read more