Unmanned Aerial Vehicles (UAVs) and Wireless sensor networks (WSNs) play an essential role in the Internet of Things (IoT) because of their easy usage and cost reduction. A UAV can used as an Air Base Station (ABS). One of the main challenges of an IoT/WSN is energy consumption, most of which relates to the transmutation part. Although many efforts have been made to propose several hybrid, dynamic, and static clustering protocols, most of these protocols apply to certain aspects of the network. So far, not many efforts have been made to present a hybrid protocol that can stay stable in various small and large-scale UAV-assisted IoT wireless networks. This paper presents a hybrid unequal hierarchy routing called an impressive clustering based on the fuzzy system (ICF). Clustering is performed at the beginning of each meta-round instead of each round to reduce overhead. This task scheduling is determined based on fuzzy systems. In ICF, a new clustering plan (CP) technique is presented that is determined based on the fuzzy system. CP is based on the scale of the network and distance of the network to the UAV whether routing should be single-hop or multi-hop. When the CP is single-hop inter-cluster routing, the cluster heads (CHs) closer to the ABS have a larger radius. In this case, CHs closer to ABS have more cluster members (CMs) to receive and data aggregation. When the CP is a multi-hop, the CHs closer to the ABS have a smaller radius. In this case, CHs closer to ABS have more energy to receive and relay data from far CHs to ABS. Choosing single-hop routing for small-scale networks helps reduce the control message's overhead. It also maintains the network's stability by selecting multi-hop routing in a large-scale network. Therefore, in different size networks, the protocol can work agreeably. Also, ICF uses the assistance to cluster heads (ACHs) technique which allows CHs to get help from some of its CMs to help share cluster load. The simulation results showed that the proposed method improves network stability, load-balancing, and network lifetime and performance.