國立虎尾科技大學 |

Implementation of Deep Convolutional Neural Network for Predicting Steering Angle in Autonomous Vehicle Systems.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Implementation of Deep Convolutional Neural Network for Predicting Steering Angle in Autonomous Vehicle Systems./
作者:	Mehdy, A K M Nuhil.
面頁冊數:	1 online resource (108 pages)
附註:	Source: Masters Abstracts International, Volume: 57-02.
標題:	Computer science. -
電子資源:	click for full text (PQDT)
ISBN:	9780355369564

Implementation of Deep Convolutional Neural Network for Predicting Steering Angle in Autonomous Vehicle Systems.
Mehdy, A K M Nuhil.

Implementation of Deep Convolutional Neural Network for Predicting Steering Angle in Autonomous Vehicle Systems. - 1 online resource (108 pages)

Source: Masters Abstracts International, Volume: 57-02.

Thesis (M.S.)--Lamar University - Beaumont, 2017.

Includes bibliographical references

Autonomous vehicle, also known as driverless, auto or self-driving vehicle is the kind of vehicle that is capable of sensing its environment, taking driving direction along with decision and navigating without human input. Having an advanced control system, an autonomous vehicle use various techniques to identify its surroundings for better navigation through the road. So far, Radar, Lidar, laser light, GPS, Odometry, and Computer Vision are the most important techniques that are being used in different ways by the engineers. Among all these robotic approaches, Computer Vision only can be the most efficient one for describing the actual surroundings of a vehicle; just like us---the human who takes all major functional decisions mostly based on the visual context in front of him.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9780355369564Subjects--Topical Terms:

573171
Computer science.
Index Terms--Genre/Form:

554714
Electronic books.

Implementation of Deep Convolutional Neural Network for Predicting Steering Angle in Autonomous Vehicle Systems.
LDR:03433ntm a2200361K 4500 001 914975
005 20180727091502.5
006 m o u
007 cr mn||||a|a||
008 190606s2017 xx obm 000 0 eng d
020 $a 9780355369564
035 $a (MiAaPQ)AAI10619631
035 $a (MiAaPQ)lamar:11090
035 $a AAI10619631
040 $a MiAaPQ $b eng $c MiAaPQ
100 1 $a Mehdy, A K M Nuhil. $3 1148597
245 1 0 $a Implementation of Deep Convolutional Neural Network for Predicting Steering Angle in Autonomous Vehicle Systems.
264 0 $c 2017
300 $a 1 online resource (108 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Masters Abstracts International, Volume: 57-02.
500 $a Adviser: Stefan Andrei.
502 $a Thesis (M.S.)--Lamar University - Beaumont, 2017.
504 $a Includes bibliographical references
520 $a Autonomous vehicle, also known as driverless, auto or self-driving vehicle is the kind of vehicle that is capable of sensing its environment, taking driving direction along with decision and navigating without human input. Having an advanced control system, an autonomous vehicle use various techniques to identify its surroundings for better navigation through the road. So far, Radar, Lidar, laser light, GPS, Odometry, and Computer Vision are the most important techniques that are being used in different ways by the engineers. Among all these robotic approaches, Computer Vision only can be the most efficient one for describing the actual surroundings of a vehicle; just like us---the human who takes all major functional decisions mostly based on the visual context in front of him.
520 $a On the other hand, due to the advances in the field of Machine Learning, computational power, and improvement in the field of digital image processing; intelligent computer vision has never been so promising, with new features and capability available. Taking all these into account, a deep learning approach, based on Convolutional Neural Network (CNN) architecture has been revealed in this thesis to take computer vision one step further that can alone help Autonomous Vehicle to both learn how to drive and to do it. A CNN is trained to map raw pixels from a single front-facing camera view directly to steering commands.
520 $a The system automatically learns internal representations of the necessary processing steps such as detecting useful road features with only the human steering angle as the training signal. Later while in testing mode, the model can predict a steering angle based on a new input image that it never seen before. Unlike typical machine learning or traditional image processing approach, this model was never been explicitly trained by manually extracted features of image, or to detect outline of roads and lanes. I have designed a novel deep neural network, based on a model proposed by NVidia Corporation in one of their papers that is related to End-To-End Learning. I have also simulated the final model by applying it to an autonomous agent that shows an autonomy value of almost 86%.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
650 4 $a Computer engineering. $3 569006
650 4 $a Automotive engineering. $3 1104081
655 7 $a Electronic books. $2 local $3 554714
690 $a 0984
690 $a 0464
690 $a 0540
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a Lamar University - Beaumont. $b Computer Science. $3 1148598
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10619631 $z click for full text (PQDT)