Terrain Correlation Correction Method for AUV Seabed Terrain Mapping

2017 ◽  
Vol 70 (5) ◽  
pp. 1062-1078 ◽  
Author(s):  
Ye Li ◽  
Teng Ma ◽  
Rupeng Wang ◽  
Pengyun Chen ◽  
Qiang Zhang
Keyword(s):  
Autonomous Underwater Vehicle ◽  
Likelihood Function ◽  
Correction Method ◽  
Sensor Nodes ◽  
Terrain Mapping ◽  
Correlation Correction ◽  
Front End ◽  
Terrain Elevation ◽  
Elevation Measurement ◽  
Self Consistency

A method is proposed for improving the accuracy and self-consistency of bathymetric maps built using an Autonomous Underwater Vehicle (AUV) to create precise prior maps for Terrain-Aided Navigation (TAN), when the Global Positioning System (GPS) or another precise location method is unavailable. This method consists of front-end and back-end. For the front-end, the AUV predicts the measurement of the bathymetry system through Terrain Elevation Measurement Extrapolation Estimation (TEMEE) and calculates the likelihood function using real measurements. After the final Inertial Navigation System (INS) error is obtained by communicating with sensor nodes, the process enters the back-end. A Terrain Correlation Correction Method (TCCM) and an Improved Terrain Correlation Correction Method (ITCCM) are proposed to solve the gradual distribution of the final INS error to each point on a path, and the accuracy of ITCCM was confirmed experimentally. Finally, a TAN simulation experiment was conducted to prove the importance and necessity of map correction using ITCCM. ITCCM was proven to be an effective and important method for correcting maps built using an AUV.

scholarly journals RGB–D terrain perception and dense mapping for legged robots

2016 ◽  
Vol 26 (1) ◽  
pp. 81-97 ◽  
Author(s):  
Dominik Belter ◽  
Przemysław Łabecki ◽  
Péter Fankhauser ◽  
Roland Siegwart
Keyword(s):  
Stereo Vision ◽  
Uncertainty Propagation ◽  
Depth Map ◽  
Mapping Method ◽  
Spatial Uncertainty ◽  
Legged Robots ◽  
Walking Robots ◽  
Terrain Mapping ◽  
Terrain Elevation ◽  
Uncertainty Models

Abstract This paper addresses the issues of unstructured terrain modeling for the purpose of navigation with legged robots. We present an improved elevation grid concept adopted to the specific requirements of a small legged robot with limited perceptual capabilities. We propose an extension of the elevation grid update mechanism by incorporating a formal treatment of the spatial uncertainty. Moreover, this paper presents uncertainty models for a structured light RGB-D sensor and a stereo vision camera used to produce a dense depth map. The model for the uncertainty of the stereo vision camera is based on uncertainty propagation from calibration, through undistortion and rectification algorithms, allowing calculation of the uncertainty of measured 3D point coordinates. The proposed uncertainty models were used for the construction of a terrain elevation map using the Videre Design STOC stereo vision camera and Kinect-like range sensors. We provide experimental verification of the proposed mapping method, and a comparison with another recently published terrain mapping method for walking robots.

scholarly journals When Speaker Recognition Meets Noisy Labels: Optimizations for Front-ends and Back-ends

2021 ◽  
Author(s):  
Lin Li ◽  
Fuchuan Tong ◽  
Qingyang Hong
Keyword(s):  
Speaker Recognition ◽  
Large Scale ◽  
Recognition Performance ◽  
Recognition System ◽  
Correction Method ◽  
Superior Performance ◽  
Label Noise ◽  
Practical Applications ◽  
Front End ◽  
Noisy Labels

A typical speaker recognition system often involves two modules: a feature extractor front-end and a speaker identity back-end. Despite the superior performance that deep neural networks have achieved for the front-end, their success benefits from the availability of large-scale and correctly labeled datasets. While label noise is unavoidable in speaker recognition datasets, both the front-end and back-end are affected by label noise, which degrades the speaker recognition performance. In this paper, we first conduct comprehensive experiments to help improve the understanding of the effects of label noise on both the front-end and back-end. Then, we propose a simple yet effective training paradigm and loss correction method to handle label noise for the front-end. We combine our proposed method with the recently proposed Bayesian estimation of PLDA for noisy labels, and the whole system shows strong robustness to label noise. Furthermore, we show two practical applications of the improved system: one application corrects noisy labels based on an utterance’s chunk-level predictions, and the other algorithmically filters out high-confidence noisy samples within a dataset. By applying the second application to the NIST SRE0410 dataset and verifying filtered utterances by human validation, we identify that approximately 1% of the SRE04-10 dataset is made up of label errors.<br>

scholarly journals Value of Information Based Data Retrieval in UWSNs

Sensors ◽  
2018 ◽  
Vol 18 (10) ◽  
pp. 3414 ◽  
Author(s):  
Fahad Khan ◽  
Sehar Butt ◽  
Saad Khan ◽  
Ladislau Bölöni ◽  
Damla Turgut
Keyword(s):  
Value Of Information ◽  
Autonomous Underwater Vehicle ◽  
Data Transfer ◽  
Autonomous Underwater Vehicles ◽  
Data Retrieval ◽  
Underwater Vehicles ◽  
Sensor Nodes ◽  
Underwater Sensor Networks ◽  
Optical Channels ◽  
Underwater Acoustic Channels

Sensor nodes in underwater sensor networks may acquire data at a higher rate than their ability to communicate over underwater acoustic channels. Autonomous underwater vehicles may mitigate this mismatch by offloading high volumes of data from the sensor nodes and ferrying them to the sink. Such a mode of data transfer results in high latency. Occasionally, these networks need to report high priority events such as catastrophes or intrusions. In such a scenario the expectation is to have a minimal end-to-end delay for event reporting. Considering this, underwater vehicles should schedule their visits to the sensor nodes in a manner that aids efficient reporting of high-priority events. We propose the use of the Value of Information metric in order to improve the reporting of events in an underwater sensor network. The proposed approach classifies the recorded data in terms of its value and priority. The classified data is transmitted using a combination of acoustic and optical channels. We perform experiments with a binary event model, i.e., we classify the events into high-priority and low-priority events. We explore a couple of different path planning strategies for the autonomous underwater vehicle. Our results show that scheduling visits to sensor nodes, based on algorithms that address the value of information, improves the timely reporting of high priority data and enables the accumulation of larger value of information.

A high-sensitivity fully passive wake-up radio front-end for wireless sensor nodes

2014 ◽  
Author(s):  
Pouya Kamalinejad ◽  
Kamyar Keikhosravy ◽  
Michele Magno ◽  
Shahriar Mirabbasi ◽  
Victor C.M. Leung ◽  
...  
Keyword(s):  
High Sensitivity ◽  
Sensor Nodes ◽  
Wireless Sensor ◽  
Wireless Sensor Nodes ◽  
Front End

Research on Improved Multidimensional Scaling Localization Algorithm for Wireless Sensor Network

2010 ◽  
Vol 44-47 ◽  
pp. 3932-3936
Author(s):  
Liang Tao ◽  
Shuai Xu ◽  
Hai Yong Chen ◽  
He Xu Xun
Keyword(s):  
Multidimensional Scaling ◽  
Search Algorithm ◽  
Correction Method ◽  
Sensor Nodes ◽  
Wireless Sensor ◽  
Localization Algorithm ◽  
Wireless Sensor Nodes ◽  
Anchor Nodes ◽  
Hardware Configuration ◽  
Positioning Algorithm

Wireless sensor networks, which are energy limited, low hardware configuration and proneness to invalidation, puts a high demand on the positioning algorithm. Therefore the improved multidimensional scaling (IMDS) algorithm is proposed. In IMDS, firstly, local positioning areas (LPA) are established by an adaptive search algorithm. So the centralized multidimensional scaling (MDS) algorithm is changed into a distributed one. Then the shortest path distances between nodes on LPA are corrected with the geometric correction method (GCM) and adjusting weight correction method (AWCM). The distances between nodes become more accurate. Finally, with information of the public nodes of LPA and anchor nodes, we get the wireless sensor nodes coordinates through coordinate transformation by the SMACOF algorithm and the classical MDS algorithm.

scholarly journals Route optimizing and following for autonomous underwater vehicle ladder surveys

2018 ◽  
Vol 15 (6) ◽  
pp. 172988141881327 ◽  
Author(s):  
Yanqing Jiang ◽  
Ye Li ◽  
Yumin Su ◽  
Ziye Zhou ◽  
Teng Ma ◽  
...  
Keyword(s):  
Autonomous Underwater Vehicle ◽  
Underwater Vehicle ◽  
Environmental Data ◽  
Confined Water ◽  
Terrain Mapping ◽  
Yaw Control ◽  
Heading Angle ◽  
Objects Detection ◽  
Transverse Position ◽  
Fuzzy Parameters

An autonomous underwater vehicle is able to conduct coverage detections, such as sea terrain mapping and submerged objects detection, using sonar. This work addresses the task of both optimizing and following routes that present a ladder shape. First, a planning method to determine a nearly optimal coverage route is designed. The track spacing is optimized considering the seabed type and the sonar range for the purpose of increasing detection probability. It also adds adaptability to confined water, such as harbors, by decomposing the geometrically concave mission region during the processing of the environmental data. Next, a decoupled and two-layered structure is adopted to design the following controller. The route is followed in the form of sequenced-lines tracking. A proportion–integral–derivative algorithm with fuzzy parameters adjustment is employed to calculate a reference heading angle according to the transverse position deviation in designing the guidance controller. An adaptive nonlinear S-surface law is adopted to design the yaw control. The route following the method is demonstrated with sonar (including side scanning sonar and multi-beam echo sounder) imagery collected in terrain mapping and object detection through sea trials.

Sign in / Sign up

Export Citation Format

Share Document