scholarly journals DANAE++: A Smart Approach for Denoising Underwater Attitude Estimation

2021 ◽  
Author(s):  
Paolo Russo ◽  
Fabiana Di Ciaccio ◽  
Salvatore Troisi
Keyword(s):  
Deep Learning ◽  
Kalman Filter ◽  
Real Time ◽  
Denoising Autoencoder ◽  
Orientation Estimation ◽  
Underwater Robots ◽  
Underwater Environment ◽  
Real Time Applications ◽  
And Performance ◽  
Vehicle Positioning

One of the main issues for underwater robots navigation is represented by the accurate vehicle positioning, which heavily depends on the orientation estimation phase. The systems employed to this scope are affected by different noise typologies, mainly related to the sensors and to the irregular noise of the underwater environment. Filtering algorithms can reduce their effect if opportunely configured, but this process usually requires fine techniques and time. This paper presents DANAE++, an improved denoising autoencoder based on DANAE, which is able to recover Kalman Filter IMU/AHRS orientation estimations from any kind of noise, independently of its nature. This deep learning-based architecture already proved to be robust and reliable, but in its enhanced implementation significant improvements are obtained both in terms of results and performance. In fact, DANAE++is able to denoise the three angles describing the attitude at the same time, and that is verified also on the estimations provided by the more performing Extended KF. Further tests could make this method suitable for real-time applications on navigation tasks.

scholarly journals DANAE++: A Smart Approach for Denoising Underwater Attitude Estimation

Sensors ◽  
2021 ◽  
Vol 21 (4) ◽  
pp. 1526
Author(s):  
Paolo Russo ◽  
Fabiana Di Ciaccio ◽  
Salvatore Troisi
Keyword(s):  
Deep Learning ◽  
Kalman Filter ◽  
Real Time ◽  
Attitude Estimation ◽  
Denoising Autoencoder ◽  
Underwater Robots ◽  
Underwater Environment ◽  
Real Time Applications ◽  
And Performance ◽  
Vehicle Positioning

One of the main issues for the navigation of underwater robots consists in accurate vehicle positioning, which heavily depends on the orientation estimation phase. The systems employed to this end are affected by different noise typologies, mainly related to the sensors and to the irregular noise of the underwater environment. Filtering algorithms can reduce their effect if opportunely configured, but this process usually requires fine techniques and time. This paper presents DANAE++, an improved denoising autoencoder based on DANAE (deep Denoising AutoeNcoder for Attitude Estimation), which is able to recover Kalman Filter (KF) IMU/AHRS orientation estimations from any kind of noise, independently of its nature. This deep learning-based architecture already proved to be robust and reliable, but in its enhanced implementation significant improvements are obtained in terms of both results and performance. In fact, DANAE++ is able to denoise the three angles describing the attitude at the same time, and that is verified also using the estimations provided by an extended KF. Further tests could make this method suitable for real-time applications in navigation tasks.

scholarly journals Real-Time Vehicle Make and Model Recognition with the Residual SqueezeNet Architecture

Sensors ◽  
2019 ◽  
Vol 19 (5) ◽  
pp. 982 ◽  
Author(s):  
Hyo Lee ◽  
Ihsan Ullah ◽  
Weiguo Wan ◽  
Yongbin Gao ◽  
Zhijun Fang
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Large Scale ◽  
Recognition Rate ◽  
Experimental Results ◽  
Learning Approach ◽  
Deep Model ◽  
Proposed Model ◽  
Real Time Applications ◽  
Model Recognition

Make and model recognition (MMR) of vehicles plays an important role in automatic vision-based systems. This paper proposes a novel deep learning approach for MMR using the SqueezeNet architecture. The frontal views of vehicle images are first extracted and fed into a deep network for training and testing. The SqueezeNet architecture with bypass connections between the Fire modules, a variant of the vanilla SqueezeNet, is employed for this study, which makes our MMR system more efficient. The experimental results on our collected large-scale vehicle datasets indicate that the proposed model achieves 96.3% recognition rate at the rank-1 level with an economical time slice of 108.8 ms. For inference tasks, the deployed deep model requires less than 5 MB of space and thus has a great viability in real-time applications.

Boosting the deep learning wavefront sensor for real-time applications [Invited]

2021 ◽  
Vol 60 (10) ◽  
pp. B119
Author(s):  
Esteban Vera ◽  
Felipe Guzmán ◽  
Camilo Weinberger
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Wavefront Sensor ◽  
Real Time Applications

An Architecture-Based Adaptation Framework for Soft Real-Time Applications

2010 ◽  
Vol 1 (4) ◽  
pp. 12-25
Author(s):  
Ning Gui ◽  
Hong Sun ◽  
Chris Blondia
Keyword(s):  
Real Time ◽  
Unified Framework ◽  
Time Component ◽  
The Face ◽  
Real Time Applications ◽  
And Performance ◽  
Time Changes ◽  
Control Application ◽  
Time Systems ◽  
Real Time Application

Real-time systems are increasingly used in dynamic changing environments with variable user needs hosting real-time applications ranging in number and nature over time. However, to the authors’ knowledge, no unified framework exists that is able to cope with those competing real-time concerns across multiple real-time application domains. This paper proposes an architecture-based framework for managing real-time components’ dependence and lifecycle in an effective and uniform way. A real-time component runtime service is proposed here to maintain the global view, control the whole lifecycle of the components, and keep existing real-time components’ promised contracts in the face of run-time changes. This framework is designed to be easily extended with other constraint resolving policies as well as dependence descriptions languages. At end of this paper, the framework is tested by a simulated control application via adaptation and performance aspects.

scholarly journals FPGA-Based Doppler Frequency Estimator for Real-Time Velocimetry

Electronics ◽  
2020 ◽  
Vol 9 (3) ◽  
pp. 456 ◽  
Author(s):  
Stefano Ricci ◽  
Valentino Meacci
Keyword(s):  
Real Time ◽  
Doppler Shift ◽  
Doppler Frequency ◽  
Low Noise ◽  
Doppler Spectrum ◽  
Frequency Interval ◽  
Field Programmable ◽  
Real Time Applications ◽  
And Performance ◽  
Frequency Estimator

In range-Doppler ultrasound applications, the velocity of a target can be measured by transmitting a mechanical wave, and by evaluating the Doppler shift present on the received echo. Unfortunately, detecting the Doppler shift from the received Doppler spectrum is not a trivial task, and several complex estimators, with different features and performance, have been proposed in the literature for achieving this goal. In several real-time applications, hundreds of thousands of velocity estimates must be produced per second, and not all of the proposed estimators are capable of performing at these high rates. In these challenging conditions, the most widely used approaches are the full centroid frequency estimate or the simple localization of the position of the spectrum peak. The first is more accurate, but the latter features a very quick and straightforward implementation. In this work, we propose an alternative Doppler frequency estimator that merges the advantages of the aforementioned approaches. It exploits the spectrum peak to get an approximate position of the Doppler frequency. Then, centered in this position, a centroid search is applied on a reduced frequency interval to refine the estimate. Doppler simulations are used to compare the accuracy and precision performance of the proposed algorithm with respect to current state of the art approaches. Finally, a Field Programmable Gate Array (FPGA) implementation is proposed that is capable of producing more than 200 k low noise estimates per second, which is suitable for the most demanding real-time applications.

scholarly journals Real-Time Automatic Investigation of Indian Roadway Animals by 3D Reconstruction Detection Using Deep Learning for R-3D-YOLOV3 Image Classification and Filtering

Electronics ◽  
2021 ◽  
Vol 10 (24) ◽  
pp. 3079
Author(s):  
Sudhakar Sengan ◽  
Ketan Kotecha ◽  
Indragandhi Vairavasundaram ◽  
Priya Velayutham ◽  
Vijayakumar Varadarajan ◽  
...  
Keyword(s):  
Deep Learning ◽  
3D Reconstruction ◽  
Real Time ◽  
Closed Loop ◽  
Multiple Views ◽  
Wild Animals ◽  
Road Accidents ◽  
Object Properties ◽  
Real Time Applications ◽  
Iot Devices

Statistical reports say that, from 2011 to 2021, more than 11,915 stray animals, such as cats, dogs, goats, cows, etc., and wild animals were wounded in road accidents. Most of the accidents occurred due to negligence and doziness of drivers. These issues can be handled brilliantly using stray and wild animals-vehicle interaction and the pedestrians’ awareness. This paper briefs a detailed forum on GPU-based embedded systems and ODT real-time applications. ML trains machines to recognize images more accurately than humans. This provides a unique and real-time solution using deep-learning real 3D motion-based YOLOv3 (DL-R-3D-YOLOv3) ODT of images on mobility. Besides, it discovers methods for multiple views of flexible objects using 3D reconstruction, especially for stray and wild animals. Computer vision-based IoT devices are also besieged by this DL-R-3D-YOLOv3 model. It seeks solutions by forecasting image filters to find object properties and semantics for object recognition methods leading to closed-loop ODT.

scholarly journals A Robust Deep Learning Approach for Position-Independent Smartphone-Based Human Activity Recognition

Sensors ◽  
2018 ◽  
Vol 18 (11) ◽  
pp. 3726 ◽  
Author(s):  
Bandar Almaslukh ◽  
Abdel Artoli ◽  
Jalal Al-Muhtadi
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Activity Recognition ◽  
Human Activity ◽  
Cost Effective ◽  
Human Activity Recognition ◽  
Position Detection ◽  
Proposed Model ◽  
Deep Convolution Neural Network ◽  
Real Time Applications

Recently, modern smartphones equipped with a variety of embedded-sensors, such as accelerometers and gyroscopes, have been used as an alternative platform for human activity recognition (HAR), since they are cost-effective, unobtrusive and they facilitate real-time applications. However, the majority of the related works have proposed a position-dependent HAR, i.e., the target subject has to fix the smartphone in a pre-defined position. Few studies have tackled the problem of position-independent HAR. They have tackled the problem either using handcrafted features that are less influenced by the position of the smartphone or by building a position-aware HAR. The performance of these studies still needs more improvement to produce a reliable smartphone-based HAR. Thus, in this paper, we propose a deep convolution neural network model that provides a robust position-independent HAR system. We build and evaluate the performance of the proposed model using the RealWorld HAR public dataset. We find that our deep learning proposed model increases the overall performance compared to the state-of-the-art traditional machine learning method from 84% to 88% for position-independent HAR. In addition, the position detection performance of our model improves superiorly from 89% to 98%. Finally, the recognition time of the proposed model is evaluated in order to validate the applicability of the model for real-time applications.

Sign in / Sign up

Export Citation Format

Share Document