Type-1 fuzzy logic controllers (FLCs) have been widely employed in many control applications as they give a good performance and it is relatively easy to extract the type-1 FLC parameters from experts. However, type-1 FLCs cannot fully handle the encountered uncertainties in changing unstructured environments as they use crisp type-1 fuzzy sets. Consequently, in order for type-1 FLCs to provide a satisfactory performance in face of high levels of uncertainties, some common practices are followed including continuously tuning the type-1 FLC or providing a set of type-1 FLCs where each FLC handles specific operation conditions. Alternatively, type-2 FLCs can handle uncertainties to give a better control performance. However, it is relatively challenging to extract from experts the footprint of uncertainty (FOU) information and consequently the type-2 fuzzy sets for type-2 FLCs. In this paper, we will present a novel method for generating the input and output type-2 fuzzy sets so that their FOUs can capture the faced uncertainties. The proposed method will generate a type-2 FLC that will try to embed the type-1 FLCs corresponding to the various operation conditions faced so far besides embedding a large number of other embedded type-1 FLCs. This will allow the type-2 FLC to handle the uncertainties trough a big number of embedded type-1 FLCs to produce a smooth and robust control performance. We will show through real world experiments how the developed type-2 FLC will handle the uncertainties and give a smooth control response that outperforms the individual and aggregated type-1 FLCs.