Augmented Reality (AR) is a technology which enhances physical environments by superimposing digital data on top of a real-world view. AR has multiple applications and use cases, bringing digital data into the physical world enabling experiences such as training staff on complicated machinery without the risks that come with such activities. Numerous other uses have been developed including for entertainment, with AR games and cultural experiences now emerging. Recently, AR has been used for developing assistive technologies, with applications across a range of disabilities. To achieve the high-quality interactions expected by users, there has been increasing integration of AR with Machine Learning (ML) algorithms. This integration offers additional functionality to increase the scope of AR applications. In this paper we present the potential of integrating AR with ML algorithms for developing assistive technologies, for the use case of locating objects in the home context.