Long-term Autonomous Mission Planning of AUV in Large-scale Complex Marine Environment

Abstract

Autonomous underwater vehicles (AUVs) have been widely used to perform marine missions. Long-term autonomous mission planning is the challenge issue that restricts their applications to the large-scale complex marine environment. In this paper, we proposed one mission planning model, which considers the various complex constraints, such as task attributes and marine environment constraints. We constructed the security situation map and formulated the mission planning into a multi-objective optimization problem that considers the time cost, the quality and quantity of completed tasks. Such a model has the extensibility features that you can associate your custom objective with extra constraints. Considering that this problem is an NP optimal problem, we used the metaheuristic algorithm -ant colony and its variants- to prove the correctness and effectiveness, and the performance of different variant algorithms are compared by the experimental simulations.