This project contains code that can be implemented on a Raspberry Pi using the Python 3 coding language. This code is designed to be implemented on an already built car with 1 Raspberry Pi, running the newest Raspian OS, as the single-board computer controlling the car. See the Objectives point below to see what the code can make the car do.
Videos On Code and Results Found - https://www.youtube.com/channel/UCKXwdkUKI9yCJQC7W0Y0AXw
- Raspberry Pi 4 Model B 2019 Quad Core 64 Bit WiFi Bluetooth (4GB)
- Raspberry Pi 4 Case w/cooling Fan and Heatsinks
- Pi Camera Wide Angled Fisheye Lens 5MP 1080P
- Micro Servo SG90 9g
- 3V DC Motors (2x)
- L298N Motor Driver
- 64 GB Micro SD Card
- Portable 3A High-Speed Power Bank
- Rechargeable 9V Batteries
- Mini Breadboard
- 470 Ohm Resistors (1x) (more to come once additional future objectives are achieved)
- Personally Designed and Printed 3D Car Body Parts (can't purchase)
- Detect lane lines
- Code: DetectLaneLines.py
- Detect stop sign and stop car
- Code: DetectStopSign_StopCar.py
- Data: stopsign_good.xml
- Train car on built track using 1 Raspberry Pi camera as the only sensor
- Code:
- To drive car from computer and collect data: DriveCar_RecordData_Threading.py
- To pull data from github and build the CNN model: BuildCNNModel.py
- Note: model.h5 and model.tflite are my models built from my data
- Code:
- Drive car from the CNN (convolutional neural network) model built based on training data in previous step
- Code: DriveAutonomously.py
Note: As these objectives are accomplished, the related code will be added to this repository and these points will be moved to the Achieved Objectives section
- Detect a traffic light at an intersection and respond accordingly
- Stop and avoid humans and other cars on crosswalks or on road
- Implement LiDAR sensor to improve human and other car interaction/path planning
- Add lights and blinkers to drive car in dark areas