Existing wireless technology and the emergence of low-cost IoT sensors in agriculture offer the possibility to gather relevant data from the soil-plant-atmosphere continuum in real time for irrigation scheduling and to have this information available remotely online. Integration of these separate technologies with automated control structures for irrigation to develop smart automated irrigation systems has the potential to reduce labour costs and increase water productivity within the cotton industry. With the final objective of developing a relatively low-cost, smart and fully automated irrigation system in collaboration with automation providers such as Padman Stops, research in this preliminary project was focused on the sensing component that such a smart automated system would have. The project showed that sensing weather, soil and crop data in real-time by means of a wireless sensor network that periodically uploads data to internet by means of low-energy and low-cost data loggers (WiField logger) offers large opportunities to optimise irrigation management through automation. In this research, data from the in-field sensors and a remotely sensed crop vegetation index (Sentinel-2 NDVI images of each bay obtained from Google Earth Engine) were ingested and analysed in Google Cloud Platform. The processed data (ETc, soil tension, etc.) plus 7-day weather forecast and soil moisture predictions were then presented in a dashboard available online using Google Data Studio in an easy to interpret manner. For the soil moisture prediction, Lasso, Decision Tree, Random Forest and Support Vector Machine modelling methods were trialled. Random Forest models gave consistently good results (mean 7-day prediction error from 8.0 to 16.9 kPa). Linear regression with two of the most important predictor variables (the square of cumulative crop evapotranspiration minus rainfall and the square of growth degree days) was not as accurate, but allowed extraction of an interpretable model. The methodology developed in this project could be used as part of a closed-loop sensing and irrigation automation system. Future research will be focused on exploring new low-cost soil and plant-based sensors, refining the algorithms developed for soil moisture prediction and the integration of such system with IoT irrigation control structures to develop a fully automated irrigation platform. With such platform, the current challenges for the widespread adoption of automated irrigation within the Australian cotton industry are likely to be significantly reduced and higher rates of adoption occur. Adoption of the proposed sensing and automation irrigation systems by cotton irrigators would result in reduced labour costs associated with irrigation water management and improvements water productivity.