An Integrated Visual Odometry System With Stereo Camera for Unmanned Underwater Vehicles

Navigation is a challenging problem in the area of underwater unmanned vehicles, due to significant electronmagnetic wave attenuation and uncertainties environments. The conventional methods, mainly implemented by acoustic devices, suffer limitations such as high cost, terrain effects low refresh rate. In this paper, novel low-cost visual navigation method, named Integrated Visual Odometry with Stereo Camera (IVO-S), has been investigated. Unlike pure odometry, proposed method fuses information from inertial sensors sonar so that it able work context-sparse practical experiments, vehicle was operated follow specific closed-loop shapes. Odoemtry Monocular (IVO-M) other popular open source SLAMs (Simultaneous Localisation Mappings), ORB-SLAM2 VINS-Mono, have used provide comparative results. cumulative error ratio quantitative evaluation analyse test It shown IVO-S sparse-feature environments accuracy, whilst also being cost solution.