Improving Simulation Realism in Developing a Fuzzy Modular Autonomous Driving System for Electric Boats

In this work an Autonomous Driving System is presented with the attempt to define a qualified standard for a specific class of electric boats. The design involved both Classical Computation and Computational Intelligence techniques (mainly Fuzzy Logic and Artificial Neural Networks) and several version of the same ADS modular architecture have obtained progressively replacing the classical modules with the intelligent ones. The modular architecture is organized in “vertical” control levels which establish a control hierarchy and “horizontal” layers that follow the direction of data processing (from real-time perception of the environment to propulsion control). Compared to what was done in our previous works, the architecture proposed for the Autonomous Driving System for a specific type of boat has been updated and it is more detailed here. Furthermore, the ad-hoc Simulator developed to test and validate the ADS was improved in realism degree and other simulations have been performed. Outcoming data and post-analyses are reported here together with our conclusions.