Abstract: The moving targets on multiobstacle sea need to avoid the obstacles, thus the motion characteristics are complicated. To improve the efficiency of satellite searching for such targets, we analyze the influence of the target’s avoidance on its probability distribution in the search map, and propose a motion prediction method for multiobstacle searching areas based on the target’s Markovian movement. Since an environmental model for moving target search is established by using the search map, we first employ the satellite detection information to update the search map, then the target transition probability is calculated using the proposed method to update the search map again. Additionally, a simulation scenario is established by adopting the maximum target’s detection probability as the satellite search strategy. The experiment’s results show the effectiveness of the method.