CMU BUGGY SIM
Simulating the Carnegie Mellon Buggy race tradition!
WHY?
CMU Buggy race has been an ages long tradition. This project aims at the implementation of modern control theory and controller design on a race simulation and was completed as part of the Modern Control Theory course work at CMU.
The preliminary idea is to gain a deeper understanding of control implementation algorithms like PID, LQR, MPC, and EKF-SLAM.
The project could prove to be extremely helpful for people competing in the buggy race every year since the racetrack data and model exactly match (although, scaled) the real world track near Flagstaff hill, CMU, Pittsburgh.
HOW?
Creating the simulation world in WEBOTS.
Measuring and generating the global track reference coordinates.
Researching, implementing (coding) the vehicle dynamics and control variables.
Coding the controller functions and files in python which will simulate the webots world model using the track reference coordinates.
Debugging the code, plotting performance plot.
Achieved a record simulation lap-time of 97.6 seconds!
Performance Plots (Control Inputs)