Multiple obstacles detection using fuzzy interface system for AUV navigation in natural water

Abstract

The detection of small moving targets, such as hazardous objects, is a challenging problem for sonar surveillance used by maritime robots' navigation in shallow water. This phenomenon is due to the ambient sea noise and reverberation. This paper presents a fuzzy algorithm to automate detection of underwater multiple objects using Mechanical Scanned Imaging Sonar (MSIS). Returned sound waves were evaluated by a fuzzy interface system (FIS). Two parameters: the Pulse length and Maximum intensity are selected as linguistic variables for the fuzzy system. The data used in this paper were collected in a natural river which is about 3 meters depth. Results demonstrate good, reliable and fast performances of the proposed algorithm.