Vehicle Dynamics and Model Characterization

Characterizing motion model of a electric-driven self-driving car, including electric motor dynamics, power train, tire dynamics, vehicle kinematics and vehicle dynamics. Since parts of parameters are unknow, we designed an adaptive filter to fuse sensors (imu,encoder) to get motion variables (longitudinal velocity, lateral velocity, yaw rate, rotation speed), collected data, and performed parameter estimation on simulink.

Model construction

Model characterization

• In order to make the this parameter estimation problem be convex, we pair several parameters to avoid nonlinearity.

  

  
  
• This result is produced by Matlab/Simulink Parameter Estimation Toolbox.