Vehicle Localization During GPS Outages With Extended Kalman Filter and Deep Learning

Liu, Jiageng; Guo, Ge

doi:10.1109/tim.2021.3097401

Cited by 73 publications

(31 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Seeing that the robustness to measurement noise is poor in complex driving environments, Liu and Guo [169] proposed an adaptive mechanism based on improved EKF and deep learning theme for vehicles, which could obtain more reliable results during GPS outages. In the maritime environment, moisture and salt spray damage the sensors quite seriously [170], so the positioning and navigation of autonomous ships at sea is full of challenges, and more intelligent positioning technology needs to be developed.…”

Section: Enhanced Localization Of Shipsmentioning

confidence: 99%

Sea-Surface Object Detection Based on Electro-Optical Sensors: A Review

Lyu

Shao

Cheng

et al. 2023

IEEE Intell. Transport. Syst. Mag.

View full text Add to dashboard Cite

Sea-surface object detection is critical for navigation safety of autonomous ships. Electrooptical (EO) sensors, such as video cameras, complement radar on board in detecting small obstacle sea-surface objects. Traditionally, researchers have used horizon detection, background subtraction, and foreground segmentation techniques to detect sea-surface objects. Recently, deep learning-based object detection technologies have been gradually applied to sea-surface object detection. This article demonstrates a comprehensive overview of sea-surface object-detection approaches where the advantages and drawbacks of each technique are compared, covering four essential aspects: EO sensors and image types, traditional object-detection methods, deep learning methods, and maritime datasets collection. In particular, sea-surface object detections based on deep learning methods are thoroughly analyzed and compared with highly influential public datasets introduced as benchmarks to verify the effectiveness of these approaches. The article also proposes the direction of future research for sea-surface object detection based on EO sensors.

show abstract

Section: Enhanced Localization Of Shipsmentioning

confidence: 99%

Sea-Surface Object Detection Based on Electro-Optical Sensors: A Review

Lyu

Shao

Cheng

et al. 2023

IEEE Intell. Transport. Syst. Mag.

View full text Add to dashboard Cite

show abstract

“…For outdoor systems, the Global Positioning System (GPS) is the most common method for localisation. 21 However, in indoor systems, many techniques such as the strength of the received signal or the time difference of arrival of two different signals are used to enable sensors to determine their locations. 22,23 In this paper, we assume the sensor nodes know their locations in the area regardless of the environment.…”

Section: Sensing Connectivity and Localisation Propertiesmentioning

confidence: 99%

A distributed self‐organising node deployment algorithm for mobile sensor networks

Ghahroudi

Shahrabi

Boutaleb

2022

Int J Communication

View full text Add to dashboard Cite

Summary Providing the appropriate coverage is quite essential for the effective functioning of many applications in Wireless Sensor Networks. Therefore, the efficiency of node deployment algorithms to supply the requested coverage is of high significance. In this paper, inspired by the equilibrium of molecules, a novel node deployment algorithm, called Smart Self‐organising Node Deployment (SSND), is proposed to provide maximum coverage. Despite other proposed algorithms, which provide coverage based on the collective movement of nodes with massive energy consumption, SSND moves one sensor in every neighbourhood at each step to reduce the sensor nodes' movement and hence the energy consumption. The chosen sensor nodes at each time step are distributedly determined by an eligibility function to reduce the non‐essential movements while improving the accuracy of the reported locations of neighbours. Our extensive simulation study shows that SSND can achieve up to 30% coverage improvement compared to those of other algorithms in most scenarios and provides an adequate trade‐off between coverage and energy consumption.

show abstract

“…When the GPS fails to provide service, the ELM will provide a predicted location. To enhance the robustness of IMU under the condition that GPS is outage, a extended Kalman filter (EKF) is proposed by [13] to eliminate the system noise. Then, a long short-term memory (LSTM) modules is proposed to predict the vehicle location.…”

Section: A Related Workmentioning

confidence: 99%

Extrinsic Information Aided Fingerprint Localization of Vehicles for Cell-Free Massive MIMO-OFDM System

Jia

Xie

et al. 2022

IEEE Open J. Commun. Soc.

View full text Add to dashboard Cite

In smart city, traffic congestion and parking problems will be solved assisted by precise vehicle location. To address the vehicle localization problem in smart city, a novel extrinsic information aided fingerprint localization algorithm is proposed in this paper. Firstly, on the basis of massive multipleinput multiple-output (MIMO) and orthogonal frequency division multiplexing (OFDM), the angle-delay domain channel matrix (ADDCM) is extracted. Then, an amplitude ratio based non-line of sight (NLoS) identification method is provided to estimate link states. For the case of LoS existence, in order to save storage space, a tuple fingerprint is proposed to record angle of LoS path. In other case, when all APs service in NLoS scenarios, to improve the localization robustness in dynamic environments, the correlated ADDCM (CADDCM) is taken as location fingerprint which can extract the constant information related to fixed scattering clusters. A greedy-based fingerprint matching scheme is used to search the nearest reference point (RP). Furthermore, the extrinsic information provided by neighbor vehicles, such as estimated location and measured distance, is utilized to improve the localization stability. Simulation results show that the extrinsic information aided algorithm could improves the localization accuracy and robustness in dynamic environments.

show abstract

Vehicle Localization During GPS Outages With Extended Kalman Filter and Deep Learning

Cited by 73 publications

References 33 publications

Sea-Surface Object Detection Based on Electro-Optical Sensors: A Review

Sea-Surface Object Detection Based on Electro-Optical Sensors: A Review

A distributed self‐organising node deployment algorithm for mobile sensor networks

Extrinsic Information Aided Fingerprint Localization of Vehicles for Cell-Free Massive MIMO-OFDM System

Contact Info

Product

Resources

About