On the Robustness of Self-Adaptive Levy-Flight

The long term goal of this research is to develop a flexible, low-cost and autonomous platform for submarine exploration. Such a platform can be used for locating submarine points of interest. Search algorithms are used to guide autonomous underwater vehicles towards their goals. In this research, a search algorithm based on Levy-flight is used. However, the effectiveness of Levy-flights depends heavily on the right choice of the parameters of the step-size distribution. Thus, a novel self-adaptation scheme was developed that automatically tunes the search parameters during the search. The simulations show that this algorithm is robust. The performance of the algorithm can be further improved by using feedback from the environment to guide the search of the autonomous underwater vehicle. The experiments demonstrate that the proposed Levy-flight search algorithm is able to improve the performance of the AUV in locating a submarine point of interest.