The use of Open-Source Boards for Data Collection and Machine Learning in Remote Deployments

Machine learning is being adopted in many walks of life to solve various problems. This is being driven by development of robust machine learning algorithms, availability of large datasets and low-cost computation resources. Some machine learning applications require deployment of devices off-the-grid for data collection and processing. Such applications require development of systems that can operate autonomously during their deployment. This paper presents how some open-source boards have been leveraged for off-grid data collection and machine learning. Advancement in technology has seen development of low-cost and low-power open-source boards that can be interfaced with a wide array of sensors for data collection and can perform computation processes. The boards are finding wide applications in data collection and machine learning initiatives. A wide array of open source boards exists in the market. The boards can generally be divided into micro controllers, single board computers and field programmable gate arrays. These boards have different properties in terms of processing capabilities, power consumption, and communication interfaces and features. For off-grid data collection and machine learning tasks, resources such as power and network for communication are limited in most cases. These factors should be considered when choosing boards for off-grid deployment tasks. The boards chosen should optimise the use of these resources while meeting the processing capabilities required for the tasks at hand.