scholarly journals Intelligent control of quad-rotor aircrafts with a STM32 microcontroller using deep neural networks

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Xiaochun Guan ◽  
Sheng Lou ◽  
Han Li ◽  
Tinglong Tang
Keyword(s):  
Neural Networks ◽  
Deep Neural Networks ◽  
Data Communication ◽  
Efficient Design ◽  
Embedded Devices ◽  
Content Type ◽  
Design Scheme ◽  
Communication Module ◽  
Aerial Vehicle ◽  
Optical Flow Sensor

Purpose Deployment of deep neural networks on embedded devices is becoming increasingly popular because it can reduce latency and energy consumption for data communication. This paper aims to give out a method for deployment the deep neural networks on a quad-rotor aircraft for further expanding its application scope. Design/methodology/approach In this paper, a design scheme is proposed to implement the flight mission of the quad-rotor aircraft based on multi-sensor fusion. It integrates attitude acquisition module, global positioning system position acquisition module, optical flow sensor, ultrasonic sensor and Bluetooth communication module, etc. A 32-bit microcontroller is adopted as the main controller for the quad-rotor aircraft. To make the quad-rotor aircraft be more intelligent, the study also proposes a method to deploy the pre-trained deep neural networks model on the microcontroller based on the software packages of the RT-Thread internet of things operating system. Findings This design provides a simple and efficient design scheme to further integrate artificial intelligence (AI) algorithm for the control system design of quad-rotor aircraft. Originality/value This method provides an application example and a design reference for the implementation of AI algorithms on unmanned aerial vehicle or terminal robots.

scholarly journals Unmanned Aerial Vehicle Visual Detection and Tracking using Deep Neural Networks: A Performance Benchmark

2021 ◽  
Author(s):  
Brian K. S. Isaac-Medina ◽  
Matt Poyser ◽  
Daniel Organisciak ◽  
Chris G. Willcocks ◽  
Toby P. Breckon ◽  
...  
Keyword(s):  
Neural Networks ◽  
Unmanned Aerial Vehicle ◽  
Deep Neural Networks ◽  
Visual Detection ◽  
Detection And Tracking ◽  
Aerial Vehicle ◽  
Performance Benchmark ◽  
A Performance

scholarly journals Deep learning or interpolation for inverse modelling of heat and fluid flow problems?

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Rainald Löhner ◽  
Harbir Antil ◽  
Hamid Tamaddon-Jahromi ◽  
Neeraj Kavan Chakshu ◽  
Perumal Nithiarasu
Keyword(s):  
Neural Networks ◽  
Heat Conduction ◽  
Deep Neural Networks ◽  
Nonlinear Behaviour ◽  
Content Type ◽  
Flow Problems ◽  
Interpolation Algorithms ◽  
Transfer Problems ◽  
First Time ◽  
Heat And Fluid Flow

Purpose The purpose of this study is to compare interpolation algorithms and deep neural networks for inverse transfer problems with linear and nonlinear behaviour. Design/methodology/approach A series of runs were conducted for a canonical test problem. These were used as databases or “learning sets” for both interpolation algorithms and deep neural networks. A second set of runs was conducted to test the prediction accuracy of both approaches. Findings The results indicate that interpolation algorithms outperform deep neural networks in accuracy for linear heat conduction, while the reverse is true for nonlinear heat conduction problems. For heat convection problems, both methods offer similar levels of accuracy. Originality/value This is the first time such a comparison has been made.

scholarly journals SIAM: Chiplet-based Scalable In-Memory Acceleration with Mesh for Deep Neural Networks

2021 ◽  
Vol 20 (5s) ◽  
pp. 1-24
Author(s):  
Gokul Krishnan ◽  
Sumit K. Mandal ◽  
Manvitha Pannala ◽  
Chaitali Chakrabarti ◽  
Jae-Sun Seo ◽  
...  
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Design Space Exploration ◽  
Deep Neural Networks ◽  
Feasible Solution ◽  
Computing System ◽  
Efficient Design ◽  
Simulation Speed ◽  
Wide Range ◽  
On Chip

In-memory computing (IMC) on a monolithic chip for deep learning faces dramatic challenges on area, yield, and on-chip interconnection cost due to the ever-increasing model sizes. 2.5D integration or chiplet-based architectures interconnect multiple small chips (i.e., chiplets) to form a large computing system, presenting a feasible solution beyond a monolithic IMC architecture to accelerate large deep learning models. This paper presents a new benchmarking simulator, SIAM, to evaluate the performance of chiplet-based IMC architectures and explore the potential of such a paradigm shift in IMC architecture design. SIAM integrates device, circuit, architecture, network-on-chip (NoC), network-on-package (NoP), and DRAM access models to realize an end-to-end system. SIAM is scalable in its support of a wide range of deep neural networks (DNNs), customizable to various network structures and configurations, and capable of efficient design space exploration. We demonstrate the flexibility, scalability, and simulation speed of SIAM by benchmarking different state-of-the-art DNNs with CIFAR-10, CIFAR-100, and ImageNet datasets. We further calibrate the simulation results with a published silicon result, SIMBA. The chiplet-based IMC architecture obtained through SIAM shows 130 and 72 improvement in energy-efficiency for ResNet-50 on the ImageNet dataset compared to Nvidia V100 and T4 GPUs.

Obtaining depth map from 2D non stereo images using deep neural networks

2018 ◽  
Vol 6 (3) ◽  
pp. 134-146
Author(s):  
Daniil Igorevich Mikhalchenko ◽  
Arseniy Ivin ◽  
Dmitrii Malov
Keyword(s):  
Neural Networks ◽  
Deep Neural Networks ◽  
Depth Map ◽  
Depth Information ◽  
Data Sets ◽  
Data Set ◽  
Content Type ◽  
Laser Sensors ◽  
Depth Images ◽  
Depth Prediction

Purpose Single image depth prediction allows to extract depth information from a usual 2D image without usage of special sensors such as laser sensors, stereo cameras, etc. The purpose of this paper is to solve the problem of obtaining depth information from 2D image by applying deep neural networks (DNNs). Design/methodology/approach Several experiments and topologies are presented: DNN that uses three inputs—sequence of 2D images from videostream and DNN that uses only one input. However, there is no data set, that contains videostream and corresponding depth maps for every frame. So technique of creating data sets using the Blender software is presented in this work. Findings Despite the problem of an insufficient amount of available data sets, the problem of overfitting was encountered. Although created models work on the data sets, they are still overfitted and cannot predict correct depth map for the random images, that were included into the data sets. Originality/value Existing techniques of depth images creation are tested, using DNN.

Knowledge-enabled digital twin for smart designing of aircraft assembly line

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Xiao Chang ◽  
Xiaoliang Jia ◽  
Kuo Liu ◽  
Hao Hu
Keyword(s):  
Design Process ◽  
Assembly Line ◽  
Design Method ◽  
Physical Interaction ◽  
Efficient Design ◽  
Content Type ◽  
Improve Design ◽  
Aircraft Assembly ◽  
Digital Twin ◽  
Design Scheme

Purpose The purpose of this paper is to provide a knowledge-enabled digital twin for smart design (KDT-SD) of aircraft assembly line (AAL) to enhance the AAL efficiency, performance and visibility. Modern AALs usually need to have capabilities such as digital-physical interaction and self-evaluation that brings significant challenges to traditional design method for AAL. The digital twin (DT) combining with reusable knowledge, as the key technologies in this framework, is introduced to promote the design process by configuring, understanding and evaluating design scheme. Design/methodology/approach The proposed KDT-SD framework is designed with the introduction of DT and knowledge. First, dynamic design knowledge library (DDK-Lib) is established which could support the various activities of DT in the entire design process. Then, the knowledge-driven digital AAL modeling method is proposed. At last, knowledge-based smart evaluation is used to understand and identify the design flaws, which could further improvement of the design scheme. Findings By means of the KDT-SD framework proposed, it is possible to apply DT to reduce the complexity and discover design flaws in AAL design. Moreover, the knowledge equips DT with the capacities of rapid modeling and smart evaluation that improve design efficiency and quality. Originality/value The proposed KDT-SD framework can provide efficient design of AAL and evaluate the design performance in advance so that the feasibility of design scheme can be improved as much as possible.

The conceptual view of unmanned aerial vehicle implementation as a mobile communication node of active data transmission network

2018 ◽  
Vol 6 (4) ◽  
pp. 174-183 ◽  
Author(s):  
Sergey V. Kuleshov ◽  
Alexandra A. Zaytseva ◽  
Alexey Y. Aksenov
Keyword(s):  
Communication Network ◽  
Unmanned Aerial Vehicle ◽  
Data Transmission ◽  
Data Communication ◽  
Group Behavior ◽  
Content Type ◽  
Group Control ◽  
Aerial Vehicle ◽  
Data Communication Network ◽  
Active Data

Purpose The purpose of this paper is to propose the basis for the unification of unmanned aerial vehicle (UAV) group control protocols for the fast deployment of communication network on territories unsuitable for stationary nodes placement. Design/methodology/approach The paper proposes the application of active data (AD) conception in which the data exist in a form of executable code allowing data packets to control its own propagation through network. The implementation is illustrated for some scenarios of UAV data communication network deployment, i.e., transmission of the AD using navigation functions and dynamic reconfiguration of the nodes. Findings The conception of AD expands the range of possible UAV group operations due to on-the-fly adaptation abilities to changes in existing or forthcoming group behavior protocols. This allows the real-time change of data transmission formats, frequency ranges, modulation types, radio network topologies which, in turn, provides the ability to dynamically form the special data transmission networks from a general purpose device temporarily reconfiguring them for data transmission task between transmitter and receiver beyond radio visibility range. Practical implications The paper includes use cases for some situation of UAV data communication network deployment. Originality/value The paper aims to expand the UAV group control principles by implementing by rapid adaptation to changes in existing or forthcoming group behavior protocols.

scholarly journals Raspberry PI for Kill Mosquitoes by Laser

2021 ◽  
Author(s):  
Ildar Rakhmatulin
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Deep Neural Networks ◽  
Raspberry Pi ◽  
Classification Problems ◽  
Power Laser ◽  
The World ◽  
Aerial Vehicle ◽  
First Time ◽  
Laser Installation

More than 700 thousand human deaths from mosquito bites are observed annually in the world. It is more than 2 times the number of annual murders in the world. In this regard, the invention of new more effective methods of protection against mosquitoes is necessary. In this article for the first time, comprehensive studies of mosquito neutralization using machine vision and a 1 W power laser are considered. Developed laser installation with Raspberry Pi that changing the direction of the laser with a galvanometer. We developed a program for mosquito tracking in real. The possibility of using deep neural networks, Haar cascades, machine learning for mosquito recognition was considered. We considered in detail the classification problems of mosquitoes in images. A recommendation is given for the implementation of this device based on a microcontroller for subsequent use as part of an unmanned aerial vehicle. Any harmful insects in the fields can be used as objects for control.

Sign in / Sign up

Export Citation Format

Share Document