Passive IR-UWB Localization System for UAV-Based Electric Facility Inspection During GPS Outage

Suh, Ui-Suk; Lee, Geunhaeng; Han, Jieun; Kim, Tae Wook; Ra, Won‐Sang

doi:10.1109/access.2023.3269285

Cited by 3 publications

(2 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another possible technology to provide navigation for the inspection robot is UWB. It is usually used to provide navigation indoors (e.g., underground substations [ 28 ]) to replace the absence of a GNSS but can also be used outdoors [ 29 , 30 ] to improve the navigation performance when GNSS signals are temporarily unavailable or have low quality. The positioning error standard deviation of the UWB systems tested on electrical facilities by different algorithmic approaches and fusion with an IMU can be lowered to a 5 cm level.…”

Section: Related Workmentioning

confidence: 99%

“…The positioning error standard deviation of the UWB systems tested on electrical facilities by different algorithmic approaches and fusion with an IMU can be lowered to a 5 cm level. Similarly, the heading estimation deviation can be reduced to ±2.5° [ 28 , 29 , 30 ]. These results are similar to the ones achieved in the other application areas [ 31 , 32 , 33 ].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

GNSS-Based Narrow-Angle UV Camera Targeting: Case Study of a Low-Cost MAD Robot

Gyrichidi,

Romanov,

Trofimov

et al. 2024

Sensors

View full text Add to dashboard Cite

One of the key challenges in Multi-Spectral Automatic Diagnostic (MAD) robot design is the precise targeting of narrow-angle cameras on a specific part of the equipment. The paper shows that a low-cost MAD robot, whose navigation system is based on open-source ArduRover firmware and a pair of low-cost Ublox F9P GNSS receivers, can inspect the 8 × 4 degree ultraviolet camera bounding the targeting error within 0.5 degrees. To achieve this result, we propose a new targeting procedure that can be implemented without any modifications in ArduRover firmware and outperforms more expensive solutions based on LiDAR SLAM and UWB. This paper will be interesting to the developers of robotic systems for power equipment inspection because it proposes a simple and effective solution for MAD robots’ camera targeting and provides the first quantitative analysis of the GNSS reception conditions during power equipment inspection. This analysis is based on the experimental results collected during the inspection of the overhead power transmission lines and equipment inspections on the open switchgear of different power plants. Moreover, it includes not only satellite, dilution of precision, and positioning/heading estimation accuracy but also the direct measurements of angular errors that could be achieved on operating power plants using GNSS-only camera targeting.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

GNSS-Based Narrow-Angle UV Camera Targeting: Case Study of a Low-Cost MAD Robot

Gyrichidi,

Romanov,

Trofimov

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

Synthetic Sensor Measurement Generation With Noise Learning and Multi-Modal Information

Romanelli,

Martinelli

2023

IEEE Access

View full text Add to dashboard Cite

Deep learning has transformed data generation, particularly in creating synthetic sensor data. This capability is invaluable in fields like autonomous driving, robotics, and computer science. To achieve this, we train models using real data, enabling them to replicate sensor data closely. These models introduce variations and noise, enhancing diversity and realism. Prominent techniques, including generative adversarial networks (GANs), variational autoencoders (VAEs), and recurrent neural networks (RNNs), excel in generating synthetic sensor data. Our paper focuses on Autoregressive Convolutional Recurrent Neural Networks (CRNN) for Multivariate Time Series Prediction. We incorporate Denoising Autoencoders (DAE) to mimic real-world noise characteristics. Our model is trained and validated using Ultra Wide Band (UWB) and Ultra High-Frequency Radio Frequency Identification (UHF-RFID) sensor data. It integrates sensor measurements and diverse information sources to produce synthetic data complementing realworld data. While demonstrated with UHF-RFID and UWB sensors, these techniques extend to industrial automation, healthcare and environmental monitoring. While our methodology exhibits broad potential, we present practical demonstrations with UHF-RFID and UWB sensors. Our deep neural network model allows researchers to construct datasets for algorithm validation, eliminating the need for costly and time-consuming data collection.

show abstract