Toward Automated Mixed Reality Interface Design: An Evolutionary Optimization Approach

In this research, we propose an innovative capsule based optimization model toward the automated user interface (UI) design of mixed reality devices. Within the model, functional, spatial, and cognitive seams unique to mixed reality devices are considered as the constraints of the capsule, the unit cohort of UI elements. A genetic evolutionary optimization solution approach was used to solve the optimization problem. The results demonstrate that the number of optimal UI elements per capsule is linearly proportional to the initial estimate of the UI elements of the capsule. The total reward of the capsule design is also linearly proportional to the number of optimal UI elements per capsule. The identified linear relationship model suggests that it is feasible to predicate the optimal number of UI elements therefore achieving the automated design of the mixed reality interface. Experimental results have validated the proposed optimization framework suggesting that the optimization reward objective function is able to match app ratings from the app store. The seam constraints of the optimization function are also consistent with the time taken to complete the task which shows the validity of using seam to model the automated mixed reality interface design.