Distant temperature and humidity monitoring: prediction and measurement

Farrukh Hafeez (Universiti Teknologi Malaysia)
Usman Ullah Sheikh (Universiti Teknologi Malaysia)
Attaullah Khidrani (Balochistan University of Engineering and Technology)
Muhammad Akram Bhayo (Quaid-e-Awam University of Engineering)
Saleh Masoud Abdallah Altbawi (Universiti Teknologi Malaysia)
Touqeer Ahmed Jumani (Mehran University of Engineering and Technology)

Indonesian Journal of Electrical Engineering and Computer Science
Vol 24, No 3: December 2021

Sensing environmental measuring parameters has a pivotal role in our everyday lives. Most of our daily life activities depend upon environmental conditions. Accurate information about these parameters also helps in several industrial applications like ventilation rate calculation, energy prediction, stock maintenance in warehouses, and saving from harmful conditions. The emergence of machine learning can make it easy to predict such time series problems. This paper describes the design of a remotely controlled robotic car for measuring and predicting humidity and temperature. A customized app for accessing the robotic car is designed to indicate predicted and realtime measured values of humidity and temperature. A sensor installed builtin helps in the measurement. The recurrent neural network (RNN) model is used to predict humidity and temperature. For this purpose, experiments are carried out in both outdoor and indoor settings. Accuracy of 85% and 90% is achieved in an outdoor environment and indoor settings.