The CSE Group of VIT CHENNAI MARKS ROVER will handle the following modules :
- Streaming of Visual Feed
- Object Detection
- GPS
- Web Interface
The Streaming module concentrates on providing live video feed from the rover.
There will be three cameras that will be used to record live video footage from the rover. These cameras will be the ARMCAM, the FRONTCAM and the REARCAM. The FRONTCAM and the REARCAM will be connected to a single Raspberry Pi and the ARMCAM will be handled only by a single pi.
The Raspberry Pi with the two cameras contains a shell script double_stream that runs when the Pi boots and activates the two cameras for video streaming. The script contains the command to start motion and execute a python script. The motion command activates the usb-camera and the pyhton script activates the Pi Camera.
The configuration in the file motion.conf are set according to the required output and the capibilities of the usb-camera. This feed is shown at the port number 8081.
The pyhton script rpi_camera_surveillance_system streams the video feed from the Pi Camera to the port number 8000.
The HTML page frame divide testing uses frame to divide the page into three parts for the feed. In each frame another HTML page is embedded which contains the feed of that particular camera.
The object detection modules concentrates on the detecting objects during the autonomous task. Objects such as blue bottle, yellow disk and red box will be present at the destination points in the autonomous task. The rover has to detect these objects and relay the confimation back to the base. Object Models are created and trained by using Machine Learning.
The GPS modules concetrates on guiding the rover to the destination points using the WGS 84 data provided. The format will be decimal degrees . The module will monitor the current coordinates and the destination coordiantes.
The Web Interface module concentrates on providing an interface to display all the data sent from the rover. It will display the current GPS coordinates of the rover, the GPS coordinates of the, the live video feed, values from the sensors and messages from the rover. Tkinter binding in python for the Tk GUI toolkit is used to create the interface. Flask server is used to deploy the web interface.