Navnet: AUV Navigation Through Deep Sequential Learning

Zhang, Xin; He, Bo; Li, Guangliang; Mu, Xiaokai; Zhou, Ying; Mang, Tanji

doi:10.1109/access.2020.2982272

Cited by 31 publications

(15 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In recent years, there has been growing interest in neural network-based underwater navigation. Various approaches have been proposed, and more recently, a research study used a deep recurrent neural network involving sequential learning with Long Short-Term Memory (LSTM) [23]. They claimed that their method outperformed Kalman-based solutions in terms of accuracy.…”

Section: Review Of Previous Workmentioning

confidence: 99%

Underwater Vehicle Positioning by Correntropy-Based Fuzzy Multi-Sensor Fusion

Shaukat

Moinuddin

Otero

2021

Sensors

View full text Add to dashboard Cite

The ability of the underwater vehicle to determine its precise position is vital to completing a mission successfully. Multi-sensor fusion methods for underwater vehicle positioning are commonly based on Kalman filtering, which requires the knowledge of process and measurement noise covariance. As the underwater conditions are continuously changing, incorrect process and measurement noise covariance affect the accuracy of position estimation and sometimes cause divergence. Furthermore, the underwater multi-path effect and nonlinearity cause outliers that have a significant impact on positional accuracy. These non-Gaussian outliers are difficult to handle with conventional Kalman-based methods and their fuzzy variants. To address these issues, this paper presents a new and improved adaptive multi-sensor fusion method by using information-theoretic, learning-based fuzzy rules for Kalman filter covariance adaptation in the presence of outliers. Two novel metrics are proposed by utilizing correntropy Gaussian and Versoria kernels for matching theoretical and actual covariance. Using correntropy-based metrics and fuzzy logic together makes the algorithm robust against outliers in nonlinear dynamic underwater conditions. The performance of the proposed sensor fusion technique is compared and evaluated using Monte-Carlo simulations, and substantial improvements in underwater position estimation are obtained.

show abstract

Section: Review Of Previous Workmentioning

confidence: 99%

Underwater Vehicle Positioning by Correntropy-Based Fuzzy Multi-Sensor Fusion

Shaukat

Moinuddin

Otero

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Given the widespread ubiquity of inertial measurement units (IMU), inertial odometry [ 18 , 44 , 46 , 53 , 77 ] is a viable alternative available to localization applications demanding small footprint, low-access delay, low-power pathway, and operating in GPS or network-denied environments. Examples of such applications include terrestrial and marine “search and rescue” missions [ 82 ], underwater sensor networks [ 15 ], oceanic biodiversity and marine health tracking [ 88 ], wildlife monitoring [ 89 ], deep-space small satellite localization [ 73 ], and localizing micro unmanned vehicles and robots [ 10 , 29 , 96 ]. For example, marine search and rescue missions [ 82 ] limit the compute device payload and resource availability (e.g., rescuers can only carry limited weight), and cannot assume continuous access to GPS or network infrastructure (e.g., the rescue operation can happen underground).…”

Section: Introductionmentioning

confidence: 99%

TinyOdom

Saha

Sandha

García

et al. 2022

Proc. ACM Interact. Mob. Wearable Ubiquitous Technol.

View full text Add to dashboard Cite

Deep inertial sequence learning has shown promising odometric resolution over model-based approaches for trajectory estimation in GPS-denied environments. However, existing neural inertial dead-reckoning frameworks are not suitable for real-time deployment on ultra-resource-constrained (URC) devices due to substantial memory, power, and compute bounds. Current deep inertial odometry techniques also suffer from gravity pollution, high-frequency inertial disturbances, varying sensor orientation, heading rate singularity, and failure in altitude estimation. In this paper, we introduce TinyOdom, a framework for training and deploying neural inertial models on URC hardware. TinyOdom exploits hardware and quantization-aware Bayesian neural architecture search (NAS) and a temporal convolutional network (TCN) backbone to train lightweight models targetted towards URC devices. In addition, we propose a magnetometer, physics, and velocity-centric sequence learning formulation robust to preceding inertial perturbations. We also expand 2D sequence learning to 3D using a model-free barometric g-h filter robust to inertial and environmental variations. We evaluate TinyOdom for a wide spectrum of inertial odometry applications and target hardware against competing methods. Specifically, we consider four applications: pedestrian, animal, aerial, and underwater vehicle dead-reckoning. Across different applications, TinyOdom reduces the size of neural inertial models by 31× to 134× with 2.5m to 12m error in 60 seconds, enabling the direct deployment of models on URC devices while still maintaining or exceeding the localization resolution over the state-of-the-art. The proposed barometric filter tracks altitude within ±0.1m and is robust to inertial disturbances and ambient dynamics. Finally, our ablation study shows that the introduced magnetometer, physics, and velocity-centric sequence learning formulation significantly improve localization performance even with notably lightweight models.

show abstract

“…In this research, an illustrated example of a three-axis, three-gimbal system was analyzed by mathematical and simulation inertial measurement unit vibration. Threeaxis degrees of freedom vector and gimbal motion data obtained by Kalman filter were provided to the graphics processing unit for offsetting each frame [10,11]. However, none of the studies mentioned above can be optimized for the efficiency of image stabilization.…”

Section: Introductionmentioning

confidence: 99%

“…The stabilization of the image only assists the system. Therefore, based on the literature [9][10][11], this study designs the input of the adaptive Kalman filter and combines the image tracking technology of [7] to improve the image quality. The goal was to use the proposed method to reduce the dependence of the optoelectronic payload driven by the motor.…”

Section: Introductionmentioning

confidence: 99%

Analysis and Design Multiple Layer Adaptive Kalman Filter Applied to Electro-Optical Infrared Payload Vision System

Lin

Yao

2022

Electronics

View full text Add to dashboard Cite

This study designs a trigger to determine the number of layers of a multi-layer adaptive Kalman filter and applies it to optoelectronic infrared payload system vision. This feature reduces the number of mechanically stabilized motors, equipment weight, CPU resources, and power for an electro-optical infrared payload system. The goal is to reduce the traditional use of multiple gyroscopes to perform calibration measurements on different gimbal frames by this design. In this study, mathematical modeling was carried out for the three-axis, three-frame camera stabilizer system, and the system foundation without motor and gimbal frame was established to achieve aperture-type camera mode. The exposure of the drone’s payload structure outside the aircraft can be reduced. This study provides the adaptive Kalman filter with the offset parameters of the camera image Minimum Output Sum of Squared Error and the three-axis degrees of freedom vector and angle data on the gyroscope. By using the image processing unit, the offset was corrected at each frame per second. The experimental results show that under the same hardware, failure limit and camera field of view constraints. The processing time by this method was compared to the traditional frame correction and full image stabilization methods. The results show that the proposed method can shorten 6 microseconds under the traditional method and can be used to provide lower power consumption, lower image delay, and a larger viewing angle range.

show abstract

Navnet: AUV Navigation Through Deep Sequential Learning

Cited by 31 publications

References 39 publications

Underwater Vehicle Positioning by Correntropy-Based Fuzzy Multi-Sensor Fusion

Underwater Vehicle Positioning by Correntropy-Based Fuzzy Multi-Sensor Fusion

TinyOdom

Analysis and Design Multiple Layer Adaptive Kalman Filter Applied to Electro-Optical Infrared Payload Vision System

Contact Info

Product

Resources

About