Skip to content

YoussefMostafaMohammed/NAVYA

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

29 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

NAVYA — Next-gen Automotive Voice and Vision Assistant

Welcome to NAVYA, a futuristic embedded system that blends vision, voice, and environmental awareness to enhance driving safety and user interaction in smart vehicles.

NAVYA transforms conventional car systems by integrating:

  • Blind spot detection & Real-time Camera mirroring via side cameras
  • Ambient light signaling for object proximity
  • Voice-activated assistant

Made for AOSP (Android Open Source Project) platforms — designed with embedded innovation in mind.


Features

Object and Blind Spot Detection System

A smart enhancement for traditional side mirrors.

  • Side Camera Integration: Mirrors are enhanced with high-definition side-mounted cameras.
  • Real-Time Object Detection:
    • Uses computer vision to detect nearby vehicles or obstacles in blind spots.
    • Draws a green circular safety indicator on the display when an object is too far.
    • Draws a yellow circular warning indicator on the display when an object is too closer.
    • Draws a red circular warning indicator on the display when an object is too close indicate there are a danger.
  • On-Screen Feedback: Drivers get intuitive visual alerts, reducing the chance of missing threats.

Ambient Light Warning System

Because your eyes shouldn't be the only way to sense danger.

  • LED Ambient Lighting Strip embedded near the display or window frame.
  • Context-Aware Color:
    • Default: Cool ambient tones.
    • Red Warning Glow: When an object approaches dangerously close, the ambient light turns red to give a clear visual cue even in peripheral vision.
  • Perfect for low-visibility or high-speed driving conditions.

Live Camera Feed Display

Mirrors are outdated. Live digital vision is here.

  • Screen Mirroring of the side cameras.
  • Built using Android’s SurfaceView or CameraX for efficient, low-latency rendering.
  • Seamless integration with AOSP UI layers.

Voice Assistant

Your car listens, understands, and talks back.

  • Wake Word Detection: Say "HiCar" to activate the assistant.
  • Voice Response:
    • Assistant greets back with "Hi, [User]".
  • Built using Android's Vosk API and Text-to-Speech (TTS) engine.
  • Potential for extended functionality: Navigation, weather updates, or media control.

Technologies & Tools

  • Android Open Source Project (AOSP)
  • TensorFlow Lite – for real-time object detection
  • CameraX / SurfaceView – for rendering side camera feeds
  • Android Vosk / TTS – for voice commands and replies
  • HAL + GPIO Control – for LED hardware integration
  • VHAL (Vehicle HAL)
  • SELinux – to enforce security policies on camera/light access

Project Structure

NAVYA/

├── app/
├── camera/
├── voice_assistant/
├── ambient_light/
├── hardware/
└── sepolicy/


Developed By

  • Abdallah Salah Mohammed
  • Aliaa Ahmed Mortada
  • Mostafa Mohammed Mohammed
  • Mousa Mahmoud Salah
  • Youssef Mostafa Mohammed
  • AOSP Project Team – Embedded Systems & Android Automotive Engineers

Future Improvements

  • SOME/IP Communication for supporting Two-Cameras
  • Send SOS message when crash happened
  • Bare-Metal Programming and CAN Bus Protocol and simulate turn signal
  • Gesture controls for touchless interaction
  • Companion mobile app for status and control

System Requirements Specification (SRS)

Functional System Requirements

ID Requirement Description
SRS-FUNC-001 The system shall detect objects in the blind spot using side-mounted cameras and computer vision.
SRS-FUNC-002 The system shall display visual safety indicators (Green/Yellow/Red) based on object proximity.
SRS-FUNC-003 The system shall mirror the side camera feeds on the digital display in real time.
SRS-FUNC-004 The system shall provide ambient light warnings using an LED strip based on object proximity.
SRS-FUNC-005 The system shall activate the voice assistant via a wake word ("HiCar").
SRS-FUNC-006 The voice assistant shall respond using TTS and perform limited interactions like greeting the user.
SRS-FUNC-007 The system shall enforce security access to camera and LED using SELinux.

Software Requirements Specification (SwRS)

Software Functional Requirements

ID Requirement Description
SWRS-FUNC-001 App shall run as a privileged system app in AOSP and use SurfaceView or CameraX for camera feed.
SWRS-FUNC-002 TensorFlow Lite shall be used for object detection in real-time on camera frames.
SWRS-FUNC-003 Vosk API shall be used to detect wake word and speech-to-text conversion.
SWRS-FUNC-004 Android TTS engine shall be used to reply back to the user.
SWRS-FUNC-005 App shall control the GPIO-based LED strip using HAL or JNI interface.
SWRS-FUNC-006 SELinux policies shall restrict access to camera and GPIO control to the NAVYA system app only.
SWRS-FUNC-007 App shall draw proximity circles using Android Canvas or OpenGL overlay.

NAVYA aims to redefine the driver experience by combining safety, intelligence, and interaction — built fully on embedded Android technologies.

About

Next-generation Automotive Voice and Vision Assistant

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • Kotlin 67.0%
  • C++ 30.7%
  • CMake 2.0%
  • Makefile 0.3%