Optimal path planning for unmanned surface vehicle using new modified local search ant colony optimization

Optimal path planning is required for unmanned surface vehicle (USV) operation. This paper proposes two new path planning methods based on ant colony algorithm considering variables such as minimum energy consumption, lowest collision probability with fixed and moving obstacles and the least travelling time. These two methods are called ant colony optimization local search (ACOLS) Out Teta and ACOLS Curve Path. In the ACOLS Out Teta, path is optimized using a new pheromone updating method based on distance with obstacles. In the ACOLS Curve Path method, an innovative way is presented to eliminate curvatures of the path, which reduces the robot’s maneuverability, decreasing the path length, energy consumption and travel time. In this paper, collision probability variable is used, providing the ability to present methods for providing an optimal path by paying a reasonable fee. Their performance is compared with each other, and with ACO and PSO.