PID Controlled Lap Sim in Simulink

This Year will be the first year Bearcats Electric Racing incorporates torque vectoring into its competition race car. In order to maximize the effectiveness of our limited testing time after the car is built and before competition, we need a way to validate the algorithm. One of the team goals for Vehicle Dynamics is to have a competition ready torque vectoring control algorithm by the time the car is ready for testing. 

Another problem we face is safely tuning PID controllers before they are implemented on the car. A poorly tuned PID can be dangerous to the driver and to spectators, and it can be time consuming and expensive to tune PIDs at the track. 

Solution

In order to be able to tune our torque vectoring system before the car is built, we need a digital twin. The main purpose of torque vectoring in an FSAE car is to improve yaw control for the driver, so the digital twin must simulate driver inputs in addition to chassis responses. This means we need two layers of controll systems, one for the driver, and one for the torque vectoring system. In order to simulate the layers of controll systems and chassis response, we decided to use simulink. Simulink has great tools for PID controllers andcontrol systems in general, it is also great for modal analysis and simulating transient suspension conditions.