Open AccessFuzzy-based potential field collision avoidance technique for unmanned surface vehicles
Author(s) -
Kantapon Tanakitkorn
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1137/1/012017
Subject(s) - collision avoidance , collision , heading (navigation) , fuzzy logic , computer science , path (computing) , potential field , obstacle , unmanned surface vehicle , field (mathematics) , function (biology) , control theory (sociology) , simulation , algorithm , aerospace engineering , artificial intelligence , physics , engineering , mathematics , marine engineering , geography , computer security , geophysics , evolutionary biology , pure mathematics , biology , archaeology , control (management) , programming language
Collision avoidance is a crucial part of the autonomous operation of an unmanned surface vehicle (USV), especially in a dynamic environment. This work presents a fuzzy-based potential field algorithm for collision prevention when a USV encounters an obstacle, either moving or static. Two types of repulsive potential functions are employed. The Gaussian repulsive potential is applied to repel the own ship (OS) away from the target ship (TS). On the other hand, the vortex potential function is used to smoothly deviate the heading of the OS towards the side that quickly resolves the collision. The fuzzy membership functions and rules are imposed to classify the encounter condition and apply the appropriate potential functions accordingly. Head-on and crossing encounter scenarios are simulated to demonstrate the performance of the proposed algorithm. The simulation results have shown that, when the collision is detected, the algorithm is effective in generating the vector field that directs the OS away from the collision path. Once the collision is resolved, the algorithm lets the OS continue the original course.