The presented Doctoral Thesis illustrates the Author’s researches in the field of VHDL based “neuro-fuzzy controllers”. The Author’s academic investigations involve numerous applications of “neuro-fuzzy controllers”, and part of Doctoral researches focuses on evaluating different implementation methods. The decision of VHDL as “controller’s hardware description language” is the outcome of the Author’s academic researches, which are the core of the Author’s international papers. Presented work starts with an overview of autonomous mobile robotics applications, automotive applications and small Autonomous Unmanned Aerial Vehicles (derivative from RC planes), which describes the context where the Doctoral Thesis is implanted. Then, the dissertation moves to the motivations behind the decision process of the selection of VHDL as “controller’s hardware description language”, strictly correlated to the flexibility and the advantages of using an FPGA instead of a multi-core MCU. A major focus is given to the FPGAs parallel processing functionality. Part of the Doctoral Thesis scrutinises methods to mitigate the complexity of a VHDL based description and the implementation of advanced learning processes. Doctoral Thesis examines a novel software tool for the high-level “neuro-fuzzy controller” description capable of executing controller simulations, optimisation tasks, performing learning/training tasks, and exporting the controller in VHDL code. The Thesis proposes an application case for the VHDL based “neuro-fuzzy controllers” researches, aiming the use of learning/training controller’s capability to off-load the mechanical design. This approach targets the controller fine-tuning through a replicable process, which shall allow adapting the controller’s parameters to the mechanical characteristics of the RC plane that shall be converted into a small Unmanned Aerial Vehicle. A series of mechanical and electrical/electronic hardware assumptions and definitions are made as pre-requisites for the controller conception. The proposal’s focus is the controller’s design strategy, scrutinising the design process, the description and the simulation of the “neuro-fuzzy controller”. Since the system’s pre-requisites and boundary conditions are finalised to deliver a general aerial vehicle controller, the Thesis aims to deliver a “neuro-fuzzy controller” capable of replicating a human being pilot behaviour. Efforts are made to establish: fuzzy controller’s simulation (fuzzy controller is the core of the “neuro-fuzzy controller” before the 5 learning/training process and the optimisation process), a learning/training process and, an optimisation process. A learning capable controller design may result in a very sophisticated design, and the designer shall rely on robust software tools; the selection of the learning/training acceleration tool becomes a crucial step of the dissertation application case. Even more important for the dissertation is the definitions of the “Simulation Conditions” on which the “core fuzzy controller” shall be tested. In fact, a mandatory condition for an appropriate learning/training process is to use a “core fuzzy controller”, already capable of performing basic tasks, as the heart of the system. What is drawn, between the lines, by the Doctoral Thesis is the introduction of a design strategy that is looking to develop solutions for complex controller architecture of mobile robotic vehicles (of any nature) or even for multiple industrial application. This work enables further investigative researches into autonomous robotics, particularly to the physical implementation of an Autonomous Aerial Unmanned Vehicle from an inexpensive RC plane. A simplified RC plane design may be used, as a worst-case scenario for the controller design, where a 3D printed homebuilt aircraft may be turned into AUAV, through the process and the algorithms disserted. Replication of the learning/training process and their iteration on different mechanics and different RC planes to be adapted into AUAV may result in information gold mining for the researchers. Indeed, the determination of reliable processes allows researchers to reutilise the same principles for totally different applications, circumscribed only by the researcher’s imagination. The Doctoral Thesis has been written in English. All summaries and conclusions and the results of the research relate to the hypothesis and the relationship between them. Researches outcome has the potential to evolve into other projects consisting of various methodologies extracted from the investigations.