Deep Learning
Advanced Driverless Car Assistance System
ADAS SYSTEM
Project URL: https://github.com/rv2442/GTSRB_Mnist_Model
Goal:
- Developing a Level 4/5 ADAS System within a Controlled Environment
- We are using a step by step approach, to reach our final goal.
- Generating Data for the Controlled Environment, Lane detection
STEP 1: LANE DETECTION & STEERING
- Research about Hardware, pre-implemented Projects, Research Papers.
- Project and Survey Report.
- Learning TensorFlow, OpenCV.
- Implementing a model on MNIST & GTSRB.
- Generating Custom Data for the Controlled Environment.
- Sampling Data from Shot Video of the 'Controlled Environment' and selectively picking up relevant data.
- Train a Model with acceptable test accuracy.
STEP 2: LANE SWITCHING
STEP 3: OBJECT DETECTION & SPEED CONTROL
STEP 4: TRAFFIC SIGN RECOGNITION
STEP 5: INTEGRATION OF ALL 4 STEPS
STEP 6: FUTURE SCOPE (CLOUD DATA ANALYTICS, CAR APP, LOCATION TRACKING)