Augmented Reality (AR) is growing rapidly and becoming a mature and robust technology, which combines virtual information with the real environment and real-time performance. It is important to ensure the acceptance and success of augmented reality systems. With the growth of elderly users, evidence shows potential trends for AR systems to support the elderly, including transport, ageing in place, entertainment and training. However, there is a lack of research to provide the theoretical framework or AR design principles to support designers when developing suitable AR applications for specific populations (e.g. older people). In my PhD thesis, I will focus on the possibility of developing and applying AR design principles to support the design of applications that address older people's requirements. In this paper, I first discuss the architecture of augmented reality and identify the relationship between different elements. Secondly, the relevant literature has been reviewed in terms of design challenges of AR and design principles. Thirdly, I formulate the five initial design principles as the fundamental work of my PhD. It is expected that design principles could help AR designers to explore quality design alternatives, which could potentially benefit the ageing population. Fourthly, I identify the AR pillbox as an example to explain how design principles can be applied to AR applications. In terms of the methodology, preparation, refinement and validation are the three main stages to achieve the research goal. Preparation stage aims to generate the preliminary AR design principles and identify the relevant scenarios that might assist the designers to understand the principles and explore the design alternatives. In the stages of refinement, a half-day workshop has been conducted to explore different design issues based on different scenarios and refine the preliminary design principles. After that, a new set of design principles will be formulated. The final stage is to validate the effectiveness of new design principles based on the previous workshop’s feedback.