A Fuzzy Logic Approach of Pareto Optimality for Multi-objective Path Planning in case of Unmanned Surface Vehicle

Abstract

Unmanned Surface Vehicles (USVs) are nowadays used in various applications for security, inspection and delivery among others. To operate in dynamic and complex environments efficiently demands an optimal path planning where multiple factors should be taken into account. In this paper, the multi-objective path planning problem of USV is formulated aiming to minimize the traveled distance maximizing in parallel the trajectory smoothness and energy efficiency. To address this multi-objective path planning problem with contradicting terms, the popular Ant Colony Optimization (ACO) algorithm is employed enhanced with the proposed Fuzzy Pareto framework. In particular, ACO is used to solve the problem by finding the Pareto solutions optimizing each single objective. Then these solutions are evaluated via the Mamdani fuzzy inference system to identify the optimal one. The ranking of the solutions is based on the defuzzification values. A case study is performed in a simulation area based on Saronic Gulf topology. The results showed that depending the needs of an operation and the conditions of the area of operations (number of obstacles, currents, and distance from the initial to the target point), each objective can impact the path quality differently.