Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 27 views

    Group Member Registration Number 01 02 03 04: Faculty of Engineering Sciences

    Uploaded by

    sharjeelhyder gopanag
    This document is a project proposal for developing a self-driving neural network car using Python and TensorFlow. The 4-member student group aims to create an AI that can assist with lane keeping and navigate to destinations using maps and GPS while avoiding collisions. The project will use computer vision, deep learning, and neural networks to allow the car to cruise without human control under simple daylight conditions. Key requirements are low cost, weight savings over LIDAR, and affordable deployment for existing vehicles. Constraints include limited hardware performance and accuracy due to time restrictions for training the AI model. The budget estimate is not included.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd

    Group Member Registration Number 01 02 03 04: Faculty of Engineering Sciences

    Uploaded by

    sharjeelhyder gopanag
    27 views6 pages
    This document is a project proposal for developing a self-driving neural network car using Python and TensorFlow. The 4-member student group aims to create an AI that can assist with lane keeping and navigate to destinations using maps and GPS while avoiding collisions. The project will use computer vision, deep learning, and neural networks to allow the car to cruise without human control under simple daylight conditions. Key requirements are low cost, weight savings over LIDAR, and affordable deployment for existing vehicles. Constraints include limited hardware performance and accuracy due to time restrictions for training the AI model. The budget estimate is not included.

    Original Description:

    sample proposal of self-driving in gta v

    Original Title

    Final Year

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document is a project proposal for developing a self-driving neural network car using Python and TensorFlow. The 4-member student group aims to create an AI that can assist with lane keeping and navigate to destinations using maps and GPS while avoiding collisions. The project will use computer vision, deep learning, and neural networks to allow the car to cruise without human control under simple daylight conditions. Key requirements are low cost, weight savings over LIDAR, and affordable deployment for existing vehicles. Constraints include limited hardware performance and accuracy due to time restrictions for training the AI model. The budget estimate is not included.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    27 views6 pages

    Group Member Registration Number 01 02 03 04: Faculty of Engineering Sciences

    Uploaded by

    sharjeelhyder gopanag
    This document is a project proposal for developing a self-driving neural network car using Python and TensorFlow. The 4-member student group aims to create an AI that can assist with lane keeping and navigate to destinations using maps and GPS while avoiding collisions. The project will use computer vision, deep learning, and neural networks to allow the car to cruise without human control under simple daylight conditions. Key requirements are low cost, weight savings over LIDAR, and affordable deployment for existing vehicles. Constraints include limited hardware performance and accuracy due to time restrictions for training the AI model. The budget estimate is not included.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    You are on page 1of 6

    GHULAM ISHAQ KHAN INSTITUTE OF ENGINEERING SCIENCES & TECHNOLOGY

    FACULTY OF ENGINEERING SCIENCES


    Final Year Project Proposal
    Group Member Registration Number
    01 Waqaruddin Rind Baloch 2016529
    (Group leader)
    02 Sharjeel Hyder 2016552

    03 Adeena Abbas 2016032

    04 Anqua Javaid Binnai-e-Tasfeen 2016078

     Previously submitted project title: NONE


    1New project title: Self Driving Neural Network Car using Tensor Flow and Python
    GHULAM ISHAQ KHAN INSTITUTE OF ENGINEERING SCIENCES & TECHNOLOGY
    FACULTY OF ENGINEERING SCIENCES
    Final Year Project Proposal

    Abstract

    The project intends to develop an AI for a simple self-driving car that stays in some lanes under simple conditions
    (High sun, clear day, no rain, no traffic...etc) in real-time and uses neural networks running on python and
    tensorflow to learn how to drive.

    The target fields on which this project is based are as follows:

    1. Programming (Python)
    2. Computer Vision (OpenCV)
    3. Deep Learning, Neural Networks (Tensor Flow)
    PROJECT GOALS AND FEATURES

    This project aims to develop following features as its standout features.

    1) Lane Assist.
    2) Using data from provided gps or maps to travel to a destination.
    3) Being able to cruise in traffic without gps or a set destination
    4) Avoiding crashes and collision warning by determining vehicle distances.

    TECHNOLOGY REQUIREMENTS

    Our proposed AI system contains multiple facets, but its modular nature makes it easy to define several key
    requirements:

     Low deployment cost for already existing vehicles.


     Easily affordable.
     Weight savings in form of reduced sensors being deployed.

    APPLICATIONS

     Cheaper alternative to LIDAR based Self-Driving Cars


     reduction of traffic accidents by eliminating human error
     make the development of self-driving cars available to more individuals and
    organizations
     assess the safety of self driving cars without requiring real world tests.
    GHULAM ISHAQ KHAN INSTITUTE OF ENGINEERING SCIENCES & TECHNOLOGY
    FACULTY OF ENGINEERING SCIENCES
    Final Year Project Proposal

    CONSTRAINTS & OBSTACLES

     Time constraints due to low performance of consumer level hardware utilized.


     High Latency experienced due to using Pascal based CUDA cores.
     Lower Level of accuracy because of insufficient time available to train AI.
     Able to drive in daylight conditions where road details are visible.
     Speed Limited to 40MPH to improve accuracy and reduce accidents.
    GHULAM ISHAQ KHAN INSTITUTE OF ENGINEERING SCIENCES & TECHNOLOGY
    FACULTY OF ENGINEERING SCIENCES
    Final Year Project Proposal

    -----------------------------------------------------------------------------------------------------

     Estimated Budget with justification:


    GHULAM ISHAQ KHAN INSTITUTE OF ENGINEERING SCIENCES & TECHNOLOGY
    FACULTY OF ENGINEERING SCIENCES
    Final Year Project Proposal

    Total Estimated Budget in PKR=____________________


    Signature of Group Leader

    Signature of Supervisor
    GHULAM ISHAQ KHAN INSTITUTE OF ENGINEERING SCIENCES & TECHNOLOGY
    FACULTY OF ENGINEERING SCIENCES
    Final Year Project Proposal
    Signature of Dean

    You might also like