GithubHelp home page GithubHelp logo

diplomski's Introduction

Diplomski

Master Thesis

Zadatak:

Opisati kako su digitalni slikovni i video signal predstavljeni na računalu. Objasniti osnovne pojmove umjetne inteligencije i strojnog učenja te opisati koncept neuronskih mreža s naglaskom na konvolucijsku neuronsku mrežu. Korištenjem programskog jezik Python i odgovarajućih modula za Python izraditi konvolucijsku neuronsku mrežu i prethodnu obradu signala. Neuronska mreža mora moći predviđati kut upravljanja. Auto se mora moći kretati samo desnom trakom i s konstantnom brzinom (15/20/25 mph). Za ostvariti konstantnu brzinu poslužiti se PID kontrolerom. Za stazu i auto iskoristiti Udacity simulator.

SAŽETAK

Zamislite da se možete sigurno i povoljno voziti s jednog odredišta na drugo dok pregledavate dokumente, gledate televiziju ili čak spavate po potrebi. Takav koncept je moguć samo ako se na tržište pojavi autonomno vozilo četvrte razine. U ovom diplomskom radu pokušavamo riješiti jedan od ključnih elemenata koje će svaki autonomni automobil trebati da prevlada, a to je izračunavanje kuta upravljanja s obzirom na cestu ispred njega. Uz određivanje kuta upravljanja, automobil mora voziti u jednoj traci što bi u našem slučaju bilo desnom trakom. Uvodni dio rada sadrži osnovne definicije te su opisani potrebni pojmovi i koncepti potrebni za praktičnu izvedbu konvolucijske neuronske mreže koja će procijeniti kut upravljanja na osnovu slike dobivene od kamere postavljene na automobilu. Upotrebom simulatora kojega je napravio Udacity omogućio je većem broju programera da se posvete izradi arhitektura umjetnih mreža kojima će se moći testirati autonomna vožnja bez velikih ulaganja i rizika za sudionike prometa. Na kraju rada se može vidjeti upotrijebljena analiza podatka i predobrada podataka zajedno s relevantnim grafovima. Štoviše, predstavljena je arhitektura konvolucijske neuronske mreže koja uspješno procjenjuje kut upravljanja dok se vozilo kreće desnom trakom.

SUMMARY

Imagine being able to safely and affordably drive from one destination to another while scrutinising documents, watching TV or even sleeping if it is required. Such a concept is only possible if a fourth level autonomous vehicle appears on the market. In this master thesis, we are trying to solve one of the key elements that every autonomous vehicle will need to overcome, which is to calculate the steering angle given the road ahead. In addition to determining the steering angle, a car must drive in one lane, which in our case would be the right lane. The introductory part of the paper provides basic definitions and describes the basic terms and concepts required for the practical implementation of a convolutional neural network that will estimate the steering angle based on the image received from a camera installed on a car. Utilizing a simulator created by Udacity has enabled more developers to devote themselves to creating artificial neural network architectures that can test autonomous driving without major investment and risk to traffic participants. At the end of the paper, we can see used data analysis and data pre-processing together with relevant graphs. Moreover, it is represented the architecture of a convolutional neural network which successfully estimating the steering angle while the vehicle is driving in the right lane.

diplomski's People

Contributors

rkal-el avatar

Watchers

James Cloos avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.