Design of a new method for occupancy monitoring in smart home care with autonomous mobile robot within Internet of Things

Abstract

The integration of autonomous mobile robots in Smart Home and their secure communication within Internet of Things with 5G networks represents a transformative shift towards more efficient, responsive, and adaptable healthcare and service delivery systems to support independent living for older people at home. This article presents a unique proposal for the possibility of implementing interoperability and secure data transmission within the communication between autonomous mobile robots and building automation technology in a Smart Home using 5G networks and also presents a novel design and application of a time-ahead $$\textrm{CO}_2\,$$ concentration prediction method for sending presence and occupancy information in monitored Smart Home Care spaces without the use of cameras to an autonomous mobile robot for time-ahead detection of deviations from the daily routine. In this study, nonlinear input-output neural network models and nonlinear autoregressive neural network model with exogenous inputs neural network models with the following best results ( $$\textrm{MSE} = 3.322 \cdot 10^{-5}, \textrm{R} = 99.913\%$$ and MAPE = 0.0565) were used. Levenberg-Marquardt algorithm, Bayes regularization algorithm and Scaled Conjugate Gradient algorithm have been used as learning algorithms. Measured waveforms of operational and technical variables for indoor environmental quality (temperature, relative humidity, light intensity and $$\textrm{CO}_2\,$$ concentration) and binary information from magnetic contacts placed on windows and doors (opening/closing of windows and doors) were used to monitor the presence of occupants in the Smart Home Care with autonomous mobile robots without the use of cameras within IoT platform with 5G networks.