Full System

Automatic Steering, Accelerating and Braking

Long/Short Range Obstacle + Traffic Signal/Sign Detection 

Autonomous Club Car

This is a 4-month individual project for the advanced Mechatronics graduate course aimed at retrofitting a club car to make it autonomous. I was responsible for the first phase of the project that ended with achieving the following objectives:

1- Automated steering (DC motor driven gear system). 

2- Automated braking and acceleration - Drive by wire (linear actuator system). 

 

3- Traffic signal recognition (camera + real-time video processing). 

 

4- Stop sign recognition (camera + real-time video processing).

 

5- Short range (ultrasonic sensors) and long range (Lidar) obstacle detection. 

The car was tested only in the lab for safety concerns and proven successful as shown in the demo videos. The next phase of the project would involve automated path generation, ground testing and finer detection through the use of GPU processors and industrial cameras (this was passed on to the next semester's students). 

 

The machine vision algorithm was run on Python using openCV. The motor and sensors were all controlled by a Microprocessor Board (Arduino) except for the LIDAR which was run by (Raspberry Pi) since fetching Lidar data takes time so there will be a delay between detection and response by the motors shall it was controlled using the same microprocessor board i.e. Arduino. The communication between Python and Arduino was through serial communication.   

Demo

© 2019 by Moustafa Raslan. All rights reserved.