In this repository, we share an implementation of preview control.

For the details of the implementation and citation please refer to :

• Self-scheduling robust preview controllers for path tracking and autonomous vehicles

    @inproceedings{boyali2017self, <br>
      title={Self-scheduling robust preview controllers for path tracking and autonomous vehicles},<br>
      author={Boyali, Ali and John, Vijay and Lyu, Zheming and Swarn, Rathour and Mita, Seichi},<br>
      booktitle={2017 11th Asian Control Conference (ASCC)},<br>
      pages={1829--1834},<br>
      year={2017},<br>
      organization={IEEE}<br>
    } 
  

The repository consists of four matlab script files.

• a file loads generic parameters for simulating the control
• a file that implements uncertain system for the LMI computations
• a post-visualizing simulation file that simulates the control performance

For computations first run a01_hinfwPoles_LMI_solution.m. For plotting the results. run a02_simulate_forrange_of_Vx.m.

In the LMI solution script, we set number of preview points and the longitudinal speed range using:

    np =50;                             % Number of Preview Points  
    VxBound=[5, 30];                     % Polytope Lower and Upper Bounds 

and compute the preview coefficients (feedforward) together with the feedback for the vehicle states. A set of figures are produced by the second script file.

Computed Gains for the number of points np=50

Tracking Performance

Pole Zero Map for the Discrete System

and BODE diagrams of the plants