Skip to content

Latest commit

 

History

History
100 lines (70 loc) · 3.1 KB

README.md

File metadata and controls

100 lines (70 loc) · 3.1 KB

Intelligent Auto - Transcending the driving hassle free.

To convert the conventional vechicle into a SDV we used sensors and measurements from the vehicle published over eCAL network and developed apps that help estimate

  1. Vehicle Stability
  2. Traffic Rules Violation Check
  3. Collison Detection
  4. Vehicle Health Monitoring
  5. Know Your Breaks

Vehicle Stability

Vehicle stability measures how stable the vehicle is on the road, this is estiamted using vehicle yaw rate, steering angle and lateral acceleration. This in turn denotes if the driver is awake.

Traffic Rules Violation Check:

This app checks for the Traffic speed limitations and Traffic warnings and alerts the user on the warnings. This app uses YOLO object dectection data to detect the traffic signals.

Collision Detection

This app checks for collison detection of vehicles such as car, trucks etc and pedestrians. A Safety region of interest is defined using the Camera's FOV and if any object comes in the ROI user will get warnings.

Vehicle Health Monitoring

This app measures the vehicle health based on the vehicle acceleration, deceleration and speed to check for jerks, quick acceleration/deceleration etc. This app is built with the idea that we can monitor various sensors in the vechile and predict a score as vehicle health - so that user would get to know about the vehicle better.

For the hackathon purpose we used existing signals and predicted a score based on user driving pattern.

Know Your Breaks

This app is enabled by a click of a button denoting that the user is going for a long drive. In the long drive, this app finds out whether the user has taken a break in between the ride for a safer journey. For example: In a 2 hour ride, driver should take rest for atleast 10 minutes for refreshments.

This is triggered by a button in the dashboard to start and stop the container

Eclipse Tools used

  • Eclipse eCAL
  • Eclipse Ankaios

Configuration

git clone https://github.com/Eclipse-SDV-Hackathon-Chapter-Two/PythonHobbits.git
cd PythonHobbits
code .

Configured container will be opened in vscode

In a seperate terminal run

ecal_play -m /measurements/<measurement_data>

To start the workload

cd scripts
start-shift2sdv

To check whether the workloads are start run

ank get workloads

To get the logs

ank- logs "<APP-NAME>"

Dashboard created

dashboard

Stability score for Funny driving scenario

funnydriving

Object detection for Collision detection scenario

funnydriving

Ankaios dashboard

ank

ROI Estimation from Camera

roi

Team members