An Integrated Visual Odometry System for Underwater Vehicles

Xu, Zhizun; Haroutunian, Maryam; Murphy, Alan J.; Neasham, Jeffrey; Norman, Rosemary

doi:10.1109/joe.2020.3036710

Cited by 12 publications

(7 citation statements)

References 43 publications

(59 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The error measured in 3D space and length of estimated trajectory in each test is presented in Table 2. The comparative results presented in the following sections are from the IVO-M [45] methods and other popular visual SLAMs or odometries, such as ORB-SLAM2, SVO, and OKVIS, are provided as well.…”

Section: Resultsmentioning

confidence: 99%

“…In [45], an Integrated Visual Odometry with a Monocular Camera (IVO-M) method has been investigated, which utilises a sonar, an IMU development kit and a gyroscope. In that method, the 3D feature points are reconstructed by the depth information from a sonar and the assumption of a partially flat seabed.…”

Section: Recent Work Of Visual Odometry and Underwater Navigationmentioning

confidence: 99%

See 1 more Smart Citation

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

Haroutunian

Murphy

et al. 2022

IEEE Access

Self Cite

View full text Add to dashboard Cite

Navigation is a challenging problem in the area of underwater unmanned vehicles, due to the significant electronmagnetic wave attenuation and the uncertainties in underwater environments. The conventional methods, mainly implemented by acoustic devices, suffer limitations such as high cost, terrain effects and low refresh rate. In this paper, a novel low-cost underwater visual navigation method, named Integrated Visual Odometry with a Stereo Camera (IVO-S), has been investigated. Unlike pure visual odometry, the proposed method fuses the information from inertial sensors and a sonar so that it is able to work in context-sparse environments. In practical experiments, the vehicle was operated to follow specific closed-loop shapes. The Integrated Visual Odoemtry with Monocular Camera (IVO-M) method and other popular open source Visual SLAMs (Simultaneous Localisation and Mappings), such as ORB-SLAM2 and VINS-Mono, have been used to provide comparative results. The cumulative error ratio is used as the quantitative evaluation method to analyse the practical test results. It is shown that the IVO-S method is able to work in underwater sparse-feature environments with high accuracy, whilst also being a low cost solution. INDEX TERMSUnderwater navigation, underwater vehicles, visual-inertial odometry, sensor fusion. sonar imaging, and wireless sensor networks at Newcastle University, in 2007, where he is currently a Senior Lecturer of communications and signal processing with the School of Electrical and Electronic Engineering. He has published over 100 conference and journal publications and his work on underwater acoustic communication and positioning has been commercialised by three U.K. companies. His research interests include underwater acoustic signal processing and device design, wireless communication networks, and biomedical instrumentation. ROSE NORMAN (Senior Member, IEEE) received the B.Eng. degree in electrical and electronic engineering from Leeds University, U.K., in 1989, and the M.Sc. and Ph.D. degrees in electrical engineering from Bradford University, U.K., in 1990 and 1994, respectively. She was the Principle Engineer at Switched Reluctance Drives Ltd., in 2004, where she is currently a Senior Lecturer with the School of Engineering. Her research interests include underwater vehicles, marine robotics and automation, and marine applications of data analytics and machine learning.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Recent Work Of Visual Odometry and Underwater Navigationmentioning

confidence: 99%

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

Haroutunian

Murphy

et al. 2022

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…An improved underwater SLAM algorithm, based on ORB features, was proposed by Lin et al [ 15 ], and the nonlinear optimization method was utilized to optimize the scale of visual odometry and the AUV pose. Xu et al combined inertial sensors, sonar and a monocular camera to predict the pose and the trajectory of the target [ 16 ]. The tank tests proved the approach was effective and reliable.…”

Section: Related Work In Underwater Visual Positioningmentioning

confidence: 99%

An Underwater Positioning System for UUVs Based on LiDAR Camera and Inertial Measurement Unit

Yang

Jia

2022

Sensors

Self Cite

View full text Add to dashboard Cite

Underwater positioning presents a challenging issue, because of the rapid attenuation of electronic magnetic waves, the disturbances and uncertainties in the environment. Conventional methods usually employed acoustic devices to localize Unmanned Underwater Vehicles (UUVs), which suffer from a slow refresh rate, low resolution, and are susceptible to the environmental noise. In addition, the complex terrain can also degrade the accuracy of the acoustic navigation systems. The applications of underwater positioning methods based on visual sensors are prevented by difficulties of acquiring the depth maps due to the sparse features, the changing illumination condition, and the scattering phenomenon. In the paper, a novel visual-based underwater positioning system is proposed based on a Light Detection and Ranging (LiDAR) camera and an inertial measurement unit. The LiDAR camera, benefiting from the laser scanning techniques, could simultaneously generate the associated depth maps. The inertial sensor would offer information about its altitudes. Through the fusion of the data from multiple sensors, the positions of the UUVs can be predicted. After that, the Bundle Adjustment (BA) method is used to recalculate the rotation matrix and the translation vector to improve the accuracy. The experiments are carried out in a tank to illustrate the effects and accuracy of the investigated method, in which the ultra-wideband (UWB) positioning system is used to provide reference trajectories. It is concluded that the developed positioning system is able to estimate the trajectory of UUVs accurately, whilst being stable and robust.

show abstract

“…The combination of multiple methods can alleviate the problems associated with the original methods but at the cost of increasing the whole system complexity, making it expensive and harder to deploy. Some systems attempt to reduce the complexity and cost, for example [34] presents the use of two low-cost inertial sensors and a sonar (altimeter). In [35], FLS images and an IMU are used with an adaptive Kalman filter to replace a DVL.…”

Section: Introductionmentioning

confidence: 99%

Sonar FoV Segmentation for Motion Estimation Using DL Networks

2022

View full text Add to dashboard Cite

Accurate navigation of an autonomous underwater vehicle is important for its reliable operation. However, this task is challenging due to limitations of radio wave propagation and poor visibility in the aquatic environment. Underwater navigation techniques based on analysing sonar images facilitated by machine learning have shown promising results. However, previously proposed techniques are still complicated for real-time applications. This paper investigates low complexity techniques for the motion estimation based on the use of images obtained by a sonar looking down to the seafloor. The sonar can use multiple beams within a field of view (FoV). Various configurations of beams are considered according to portions of the FoV covered and two estimation approaches are investigated. In one approach, the sonar images are directly processed by a deep learning (DL) network, whereas in the other, the images are converted into (reduced size) vectors before applying them to a DL network. The vector approach shows a significantly lower computation time (about 10 times faster), which makes it suitable for realtime applications. Both the approaches show a similar estimation accuracy, about 10% of the maximum magnitude of the motion. The vector technique has been used to estimate a simulated trajectory and compare the estimate with the ground truth, which showed a good match. It has also been applied to estimate the trajectory of an imaging sonar from a real data set from a ship's hull inspection. The estimated trajectory has successfully been used to build a mosaic by merging the sonar images from the real data set.

show abstract

An Integrated Visual Odometry System for Underwater Vehicles

Cited by 12 publications

References 43 publications

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

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

An Underwater Positioning System for UUVs Based on LiDAR Camera and Inertial Measurement Unit

Sonar FoV Segmentation for Motion Estimation Using DL Networks

Contact Info

Product

Resources

About