scholarly journals A Review on IoT Deep Learning UAV Systems for Autonomous Obstacle Detection and Collision Avoidance

2019 ◽  
Vol 11 (18) ◽  
pp. 2144 ◽  
Author(s):  
Paula Fraga-Lamas ◽  
Lucía Ramos ◽  
Víctor Mondéjar-Guerra ◽  
Tiago M. Fernández-Caramés
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Collision Avoidance ◽  
Large Scale ◽  
Obstacle Detection ◽  
Unmanned Aerial Systems ◽  
Promising Alternative ◽  
Communication Architectures ◽  
Recent Developments ◽  
Aerial Systems

Advances in Unmanned Aerial Vehicles (UAVs), also known as drones, offer unprecedented opportunities to boost a wide array of large-scale Internet of Things (IoT) applications. Nevertheless, UAV platforms still face important limitations mainly related to autonomy and weight that impact their remote sensing capabilities when capturing and processing the data required for developing autonomous and robust real-time obstacle detection and avoidance systems. In this regard, Deep Learning (DL) techniques have arisen as a promising alternative for improving real-time obstacle detection and collision avoidance for highly autonomous UAVs. This article reviews the most recent developments on DL Unmanned Aerial Systems (UASs) and provides a detailed explanation on the main DL techniques. Moreover, the latest DL-UAV communication architectures are studied and their most common hardware is analyzed. Furthermore, this article enumerates the most relevant open challenges for current DL-UAV solutions, thus allowing future researchers to define a roadmap for devising the new generation affordable autonomous DL-UAV IoT solutions.

scholarly journals VIRTUALIZING SUPER-COMPUTATION ON-BOARD UAS

2015 ◽  
Vol XL-7/W3 ◽  
pp. 1291-1298 ◽  
Author(s):  
E. Salami ◽  
J. A. Soler ◽  
R. Cuadrado ◽  
C. Barrado ◽  
E. Pastor
Keyword(s):  
Real Time ◽  
High Performance ◽  
Unmanned Aerial Systems ◽  
Sensing Applications ◽  
Recent Developments ◽  
Level Information ◽  
Information Products ◽  
High Level ◽  
The Cost ◽  
Aerial Systems

Unmanned aerial systems (UAS, also known as UAV, RPAS or drones) have a great potential to support a wide variety of aerial remote sensing applications. Most UAS work by acquiring data using on-board sensors for later post-processing. Some require the data gathered to be downlinked to the ground in real-time. However, depending on the volume of data and the cost of the communications, this later option is not sustainable in the long term. This paper develops the concept of virtualizing super-computation on-board UAS, as a method to ease the operation by facilitating the downlink of high-level information products instead of raw data. Exploiting recent developments in miniaturized multi-core devices is the way to speed-up on-board computation. This hardware shall satisfy size, power and weight constraints. Several technologies are appearing with promising results for high performance computing on unmanned platforms, such as the 36 cores of the TILE-Gx36 by Tilera (now EZchip) or the 64 cores of the Epiphany-IV by Adapteva. The strategy for virtualizing super-computation on-board includes the benchmarking for hardware selection, the software architecture and the communications aware design. A parallelization strategy is given for the 36-core TILE-Gx36 for a UAS in a fire mission or in similar target-detection applications. The results are obtained for payload image processing algorithms and determine in real-time the data snapshot to gather and transfer to ground according to the needs of the mission, the processing time, and consumed watts.

scholarly journals Deep Learning-Based Object Detection for Unmanned Aerial Systems (UASs)-Based Inspections of Construction Stormwater Practices

Sensors ◽  
2021 ◽  
Vol 21 (8) ◽  
pp. 2834
Author(s):  
Billur Kazaz ◽  
Subhadipto Poddar ◽  
Saeed Arabi ◽  
Michael A. Perez ◽  
Anuj Sharma ◽  
...  
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
Pollution Prevention ◽  
Aerial Images ◽  
Unmanned Aerial Systems ◽  
Construction Sites ◽  
State Regulations ◽  
Novel Approach ◽  
Model Training ◽  
Aerial Systems

Construction activities typically create large amounts of ground disturbance, which can lead to increased rates of soil erosion. Construction stormwater practices are used on active jobsites to protect downstream waterbodies from offsite sediment transport. Federal and state regulations require routine pollution prevention inspections to ensure that temporary stormwater practices are in place and performing as intended. This study addresses the existing challenges and limitations in the construction stormwater inspections and presents a unique approach for performing unmanned aerial system (UAS)-based inspections. Deep learning-based object detection principles were applied to identify and locate practices installed on active construction sites. The system integrates a post-processing stage by clustering results. The developed framework consists of data preparation with aerial inspections, model training, validation of the model, and testing for accuracy. The developed model was created from 800 aerial images and was used to detect four different types of construction stormwater practices at 100% accuracy on the Mean Average Precision (MAP) with minimal false positive detections. Results indicate that object detection could be implemented on UAS-acquired imagery as a novel approach to construction stormwater inspections and provide accurate results for site plan comparisons by rapidly detecting the quantity and location of field-installed stormwater practices.

scholarly journals Computer vision based obstacle detection and target tracking for autonomous vehicles

2021 ◽  
Vol 336 ◽  
pp. 07004
Author(s):  
Ruoyu Fang ◽  
Cheng Cai
Keyword(s):  
Neural Network ◽  
Computer Vision ◽  
Deep Learning ◽  
Target Tracking ◽  
Real Time ◽  
Autonomous Vehicles ◽  
Autonomous Vehicle ◽  
Obstacle Detection ◽  
Pid Algorithm ◽  
Deep Learning Neural Network

Obstacle detection and target tracking are two major issues for intelligent autonomous vehicles. This paper proposes a new scheme to achieve target tracking and real-time obstacle detection of obstacles based on computer vision. ResNet-18 deep learning neural network is utilized for obstacle detection and Yolo-v3 deep learning neural network is employed for real-time target tracking. These two trained models can be deployed on an autonomous vehicle equipped with an NVIDIA Jetson Nano motherboard. The autonomous vehicle moves to avoid obstacles and follow tracked targets by camera. Adjusting the steering and movement of the autonomous vehicle according to the PID algorithm during the movement, therefore, will help the proposed vehicle achieve stable and precise tracking.

Leveraging coherent wave field analysis and deep learning in fiber-optic seismology

2021 ◽  
Author(s):  
Benjamin Schwarz ◽  
Korbinian Sager ◽  
Philippe Jousset ◽  
Gilda Currenti ◽  
Charlotte Krawczyk ◽  
...  
Keyword(s):  
Deep Learning ◽  
Wave Field ◽  
Real Time ◽  
Learning Strategies ◽  
Data Storage ◽  
Large Scale ◽  
Fiber Optic ◽  
Positive Impact ◽  
Field Analysis ◽  
Coherent Wave

<p><span>Fiber-optic cables form an integral part of modern telecommunications infrastructure and are ubiquitous in particular in regions where dedicated seismic instrumentation is traditionally sparse or lacking entirely. Fiber-optic seismology promises to enable affordable and time-extended observations of earth and environmental processes at an unprecedented temporal and spatial resolution. The method’s unique potential for combined large-N and large-T observations implies intriguing opportunities but also significant challenges in terms of data storage, data handling and computation.</span></p><p><span>Our goal is to enable real-time data enhancement, rapid signal detection and wave field characterization without the need for time-demanding user interaction. We therefore combine coherent wave field analysis, an optics-inspired processing framework developed in controlled-source seismology, with state-of-the-art deep convolutional neural network (CNN) architectures commonly used in visual perception. While conventional deep learning strategies have to rely on manually labeled or purely synthetic training datasets, coherent wave field analysis labels field data based on physical principles and enables large-scale and purely data-driven training of the CNN models. The shear amount of data already recorded in various settings makes artificial data generation by numerical modeling superfluous – a task that is often constrained by incomplete knowledge of the embedding medium and an insufficient description of processes at or close to the surface, which are challenging to capture in integrated simulations.</span></p><p><span>Applications to extensive field datasets acquired with dark-fiber infrastructure at a geothermal field in SW Iceland and in a town at the flank of Mt Etna, Italy, reveal that the suggested framework generalizes well across different observational scales and environments, and sheds new light on the origin of a broad range of physically distinct wave fields that can be sensed with fiber-optic technology. Owing to the real-time applicability with affordable computing infrastructure, our analysis lends itself well to rapid on-the-fly data enhancement, wave field separation and compression strategies, thereby promising to have a positive impact on the full processing chain currently in use in fiber-optic seismology.</span></p>

Development and Flight Validation of Proposed Unmanned Aerial System Handling Qualities Requirements

2021 ◽  
Author(s):  
Christina M. Ivler ◽  
Kevin Truong ◽  
Declan Kerwin ◽  
Joel Otomize ◽  
Danielle Parmer ◽  
...  
Keyword(s):  
Large Scale ◽  
Performance Standards ◽  
Unmanned Aircraft ◽  
Performance Standard ◽  
Great Promise ◽  
Unmanned Aerial Systems ◽  
Dynamic Scaling ◽  
Scoring Method ◽  
Handling Qualities ◽  
Aerial Systems

Unmanned aerial systems, commonly known as drones, present new opportunities to perform autonomous tasks. Handling qualities requirements for manned vertical lift aircraft have been well defined and documented. The need to define handling qualities requirements for vertical take-off and landing (VTOL) unmanned aerial systems (UAS) to meet mission demands is of paramount importance for all potential operators and procurement agencies. One way to relate handling qualities specifications of large-scale manned and subscale unmanned aircraft is through Froude dynamic scaling. Froude dynamic scaling based on hub-to-hub distance has shown great promise in relating the natural frequencies of scaled multicopters. There have been recent efforts to develop a VTOL-UAS handling qualities standard by scaling mission task elements and rating their performance through a Trajectory, Tracking, and Aggression (TTA) score. This paper proposes a new performance standard adapted from the TTA scoring method, along with a modified Cooper–Harper scale as a VTOL-UAS handling qualities framework that is consistent with the spirit of Aeronautical Design Standard 33 (ADS-33). These newly proposed performance standards were then validated through simulation and flight testing on a small hexacopter UAS, flown at the University of Portland. A key outcome of this work is the flight verification of a key dynamic response metric, the disturbance rejection bandwidth, and associated validation of Froude scaling for predicted handling qualities metrics.

scholarly journals Progressive System: A Deep-Learning Framework for Real-Time Data in Industrial Production

Processes ◽  
2020 ◽  
Vol 8 (6) ◽  
pp. 649
Author(s):  
Yifeng Liu ◽  
Wei Zhang ◽  
Wenhao Du
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Large Scale ◽  
Quality Data ◽  
Time Data ◽  
High Quality ◽  
Real Time System ◽  
High Quality Data ◽  
Learning Framework ◽  
Data Accumulation

Deep learning based on a large number of high-quality data plays an important role in many industries. However, deep learning is hard to directly embed in the real-time system, because the data accumulation of the system depends on real-time acquisitions. However, the analysis tasks of such systems need to be carried out in real time, which makes it impossible to complete the analysis tasks by accumulating data for a long time. In order to solve the problems of high-quality data accumulation, high timeliness of the data analysis, and difficulty in embedding deep-learning algorithms directly in real-time systems, this paper proposes a new progressive deep-learning framework and conducts experiments on image recognition. The experimental results show that the proposed framework is effective and performs well and can reach a conclusion similar to the deep-learning framework based on large-scale data.

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.

Sign in / Sign up

Export Citation Format

Share Document