The majority of Human--Machine Interfaces (HMIs) designed for the tele-operation of unmanned vehicles only present information visually. Frequent overloading of the operator's visual channel may result in accidents. To reduce the occurrence of accidents and improve overall operating performance, this thesis proposes the implementation of the audio Augmented Reality (AR) paradigm in the form of a spatial auditory display for complex navigation tasks such as path following (route guidance) or trajectory tracking. Therefore, the first novelty of this thesis is the audio AR for human-in-the-loop systems that improves the efficiency of the operator in high visual workload scenarios. Another novelty of the thesis is development of the virtual target based guidance method for complex guidance tasks. The thesis proposes the application of a modified ``lookahead distance" guidance strategy, as a guidance law for generating a reference. The reference is presented to the operator as a spatial auditory image of the virtual target to be followed. The result is the second contribution of the thesis: dynamic virtual target guidance method suitable for complex guidance tasks of unmanned remotely operated marine vehicles. Before being presented to the operator, the reference information is preconditioned through ``supernormal" transformations to increase azimuth and distance resolution in regions of interest and thus to improve efficiency and spatial acuity, the comparative disadvantages of the hearing channel. The third contribution of the thesis is the reference remapping transformation, adjusted for human hearing, enhancing the spatial hearing resolution in the selected area of interest and improving guidance of remotely operated marine vehicles. A series of experiments show that the proposed guidance strategy provides a comprehensible and effective auditory reference. The enhanced supernormal auditory interface improves performance by reducing the tracking error of both path following and trajectory tracking compared to a non-enhanced interface without the supernormal transformation. Experimental data confirm that under high visual and cognitive load, auditory interface emerges as an important advantage. It was shown that especially in high-load operational environment unloading the operator yields improved navigation quality. Experiment participants report that the novel auditory interface is intuitive and easy to use, providing very good situational awareness. Presented results clearly show that there is a rationale for using AR paradigm (visual and audio) for the purpose of ROV navigation. While the auditory interface investigated in this paper is shown to be effective for the tele-operation of unmanned vehicles, it may have other applications suc