Abstract
<p>The efficiency of locating a target by autonomous underwater vehicles (AUVs) depends on the selected search strategy. The selected search strategy should take into consideration all aspects of the nature and the behaviour of the search agent, which executes the search. It should also take into account the nature of the search environment. The long-term goal of this research is to use a small cooperative swarm of AUVs to locate a phenomenon of interest in a predefined marine environment. In this paper, some characteristics of search models that can promote constructive collaboration between AUVs for effective search are discussed. Two heuristic algorithms to locate a target are proposed. These heuristics do not use the exact location of the AUV to decide on the next search action as a mean to alleviate the impact of the localisation problem. The two heuristics outperform blind search algorithms in a search environment with two targets of different priorities. The impact of the localisation problem is considered by evaluating the performance of the algorithms in the presence of localisation errors. The results show that one of the proposed informed search heuristics is able to improve the search performance even in the presence of localisation errors.</p>