Development of indoor CO2 monitoring system with an integrated fixed sensor and a mobile measuring robot

Abstract This paper proposes an advanced system for optimizing indoor environmental quality (IEQ) in office environments that integrates fixed sensors with a mobile measuring robot (MMR). A demand-based measurement strategy that uses human detection and predictive analytics via machine learning is used to enhance data collection accuracy and efficiency. The system incorporates voice notifications to prompt occupants to perform actions that improve IEQ. The MMR's operational capabilities and coordination with fixed sensors allow the system to achieve high precision and efficiency in office environments. The system's effectiveness is validated through empirical studies (two preliminary experiments and two main experiments) in real office settings. The first preliminary experiment identified measurement blind spots and the second preliminary experiment tested the equivalence of environmental measurements between the MMR and fixed sensors. The first main experiment showed the system's human detection function for efficient and precise environmental measurement and the voice notification function for prompting occupants to perform actions that improve IEQ. The second main experiment showed the system's predictive accuracy in forecasting CO2 levels using neural network models. The main experiments demonstrate that the system can effectively guide MMR operations, reduce measurement times, and accurately predict environmental changes. The proposed system is a comprehensive solution for IEQ enhancement in office buildings.