The attack planning of vessel targets is related to complex multi-objective optimization characterized by difficult convergence and numerous local optima. Traditional optimization algorithms struggle with slow convergence and are prone to getting stuck in local optima， making it challenging to find optimal solutions. This paper proposed a vessel target attack planning method based on a multi-objective marine predator algorithm with a hybrid update strategy. By improving the population initialization method， constructing an optimization phase division strategy based on motion state coefficients， and establishing a hybrid motion mode， this approach enhanced the exploration and exploitation capabilities in each optimization phase of the vessel target attack planning solution. This improved the global optimization performance， ultimately obtaining the optimal vessel target attack planning. Tests conducted in three different scenarios demonstrated that the proposed multi-objective marine predator algorithm with a hybrid update strategy offered superior optimization performance， enabling more accurate planning for vessel target attack.