文献类型：会议论文

英文题名：A Low-Cost Visual Inertial Odometry System for Underwater Vehicles

作者：Xu, Zhizun[1]; Haroutunian, Maryam[2]; Murphy, Alan J[2]; Neasham, Jeff[2]; Norman, Rose[2]

机构：[1] Guangdong Ocean University, Maritime College, Guangdong Province, Zhanjiang, China; [2] Newcastle University, School of Engineering, Newcastle upon Tyne, United Kingdom

会议论文集：2021 4th International Conference on Mechatronics, Robotics and Automation, ICMRA 2021

会议日期：October 22, 2021 - October 24, 2021

会议地点：Virtual, Zhanjiang, China

语种：英文

外文关键词：Costs - Vehicles

外文摘要：The localization is a crucial issue for underwater vehicles. In the paper, a lightweight visual-inertial odometry is proposed. With dual inertial sensors giving the information of acceleration and attitude, an optical camera providing the seabed images where feature points are tracked by an optical flow algorithm, linear motion of the vehicle can be obtained by computing coordinate transformations and in the fusion section, the control input is also considered. The computational complexity of the proposed method is reduced dramatically relative to other methodologies, and the optical flow algorithm can allow the system to work in poor context environment conditions. The results evaluated by practical experiments show that the method is an effective, low-cost solution for underwater localization. ? 2021 IEEE.