MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202641110502 A) filed by Vinitha K; Saromi N; Mohammed Burhan Ahmed; C. Ushapriya; Megha Deelip Kokate; L. Steffina Morin; Mrs Jyoti Rahul Gaikwad; Jeevitha S; Mayuri R Gadhe; Fazil A. Sheikh; N. Anandha Priya; and Dr Pramod Gupta on September 15, 2026, for Iot-Enabled Machine Learning System For Real-Time Air Pollution Prediction And Adaptive Control In Smart Cities.
Inventors include Vinitha K; Saromi N; Mohammed Burhan Ahmed; C. Ushapriya; Megha Deelip Kokate; L. Steffina Morin; Mrs Jyoti Rahul Gaikwad; Jeevitha S; Mayuri R Gadhe; Fazil A. Sheikh; N. Anandha Priya; and Dr Pramod Gupta.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: The present invention relates to an IoT-Enabled Machine Learning System for Real-Time Air Pollution Prediction and Adaptive Control in Smart Cities. Existing smart city air quality systems use connected sensors and digital monitoring platforms to collect information about particulate matter, gaseous pollutants, weather conditions, traffic, and other environmental factors. Such systems are useful for observing pollution levels, but many mainly report present conditions or generate predictions without directly connecting those predictions with suitable local control actions. They may also rely on individual sensor readings or fixed limits and may not adequately reflect changing pollution behaviour at different locations. The present invention addresses these limitations through an integrated system comprising an IoT-Based Air Quality Data Acquisition Module, Air Pollution Feature Engineering Module, Machine Learning-Based Pollution Prediction Module, Location-Based Pollution Risk Assessment Module, Adaptive Pollution Control and Decision Support Module, Smart City Monitoring and Alert Module, and Continuous Learning and Feedback Module. The system generates Air Quality Index Score (AQIS), Particulate Pollution Load Index (PPLI), Pollution Exposure Risk Score (PERS), Environmental Dispersion Factor (EDF), and Pollution Trend Index (PTI) from current and historical information. These measures are used with machine learning to predict present and near-future pollution conditions for individual locations. Based on the predicted condition, the system can recommend or support suitable traffic management, emission reduction, ventilation adjustment, public warning, or other control actions. Subsequent sensor readings and control outcomes are fed back to improve future predictions and decisions. The invention can enable earlier pollution warnings, location-specific risk identification, more responsive control, improved environmental monitoring, and continuous adaptation to changing urban air quality conditions. FIG.1
Disclaimer: Curated by HT Syndication.
