Improved routing in dynamic environments with moving obstacles using a hybrid Fuzzy-Genetic algorithm

Abstract

Routing, as one of the important problems in the field of robotics, is a complicated issue in real and dynamic environments. In this study, routing was simulated using Genetic algorithm and Fuzzy logic. It was observed that the time consumed for reaching the destination in Fuzzy logic was much less than the time spent in Genetic algorithm. Furthermore, the distance traveled by Genetic algorithm was less than the distance obtained from routing by Fuzzy logic. Therefore, to determine the optimal path of motion, the hybrid Fuzzy-Genetic method was used. In Fuzzy logic, distance from the nearest obstacle and the angle difference with the target node were selected as the two node-to-node routing criteria. To reduce the traveled distance in the Fuzzy method, the Genetic algorithm was used to optimally adjust the Fuzzy rules table. In the simulation, the proposed method showed a relatively better performance than both mentioned algorithms in terms of distance and time. In the best case, the traveled distance from origin to destination in the hybrid Fuzzy-Genetic method was reduced by 32% compared to Fuzzy logic and the consumed time to reach the destination was reduced by 43% compared to Genetic algorithm.