Design of Combined Neural Network and Fuzzy Logic Controller for Marine Rescue Drone Trajectory-Tracking

In recent years, the research on drones has increased rapidly because of its high applicability in many fields and its great development potential. In the maritime sector too, especially marine rescue, a Drone with a compact size and fast flight speed is an effective solution in search and surveillance, giving quick results and being very convenient. When operating at sea, marine rescue drones are often affected by the environment, especially wind, which leads to turbulence that causes the drone to deviate from its predetermined flight trajectory. To overcome the above problem, the author has proposed the application of a Neural-Fuzzy controller for unmanned marine rescue aircraft presented in this paper introduces a controller that combines neural networks and fuzzy controllers that enhance the efficiency of the drone's trajectory tracking. The paper presents the mathematics of a quadcopter described by the Newton-Euler equations. Presentation on stable flight control and trajectory control of marine rescue drones. In this paper, Matlab/Simulink is used to describe the operation of the Drone, and the characteristics obtained after using the simulation are used to compare, test, and analyze the system. The obtained results show that the Neural-Fuzzy controller is much more sensitive, more resistant to turbulence, and can be used on different sizes, weights, and configurations of drones without adjusting PID gain.

Automated summary

This paper introduces the application of a neural-fuzzy controller in marine rescue drones to enhance trajectory tracking efficiency, and conducts simulations using Matlab/Simulink for validation.