planning a controller which calculates in real time the optimal trajectory for the vehicle and outputs the commands for the vehicle in order to drive along the trajectory using MPC.
MPC is a multivariable control algorithm based on an iterative, finite-horizon model. MPC consists of: Dynamic model and Optimization cost function
steps:
- Implementing the Pure Pursuit algorithm using a simulation based on the bicycle model and understanding the limits of the Pure Pursuit algorithm
- Using the algorithm on a real vehicle under the limitations found in the first step
- Upgrading to MPC algorithm and calibrating in order to get maximal results