An improved ant colony optimization algorithm for unmanned surface vehicle local path planning with multi-modality constraints

Abstract

The increase application of machine learning and artificial intelligence in the field of robotics, have emerged the need for real time algorithms for autonomous unmanned aircraft and surface vehicles. In recent years, various studies have been conducted for path planning of unmanned surface vehicle (USV), especially to maritime transportation. Most of the traditional approaches include optimization algorithms for finding the shortest or fastest path. However, USV motion in complex environments demand multi-objective optimization and multi-modality constraints to cope with dynamic environments and moving obstacles. To this end, an improved Ant Colony Optimization with Fuzzy Logic (ACO-FL) is proposed to deal with local path planning for obstacle avoidance by taking into account wind, current, wave and dynamic obstacles. The proposed algorithm was compared to original ACO and popular Bug algorithm in simulation tests. The results showed that ACO-FL reached better performance compared to the other algorithms under examination in terms of optimal solution and convergence speed. Thus, the proposed algorithm can be considered as an effective approach for path planning of USVs.