vehiclecontrol_gt

Repository for Game Theory Work related to the IDSC Gokart Lab

Folder Content

  1. ForcesKinematicVersion_GameTheory (Developer: Enrico Mion)
  2. ForcesMPCPathFollowingDynamic_GameTheory (Developer: Enrico Mion)
  3. Manual_ibr (Developer: Thomas Andrews)
  4. shared_dynamic (Developer: Enrico Mion)
  5. Tire_analysis (Developer: Enrico Mion)

How to use the folder:

  1. Get a Forces Licence and download Forces.

  2. Open ForcesKinematicVersion_GameTheory Folder content:

A) Animation : contains files that draw animation of the vehicles

B) auxiliary_function : contains some auxiliary functions about torque vectoring

C) casadi : auxiliary functions about spline computation

D) c files : files for the go kart

E) constraints : contains the files in which you can define non linear constraints

F) draw_files : contains files to draw simulation results

G) models : contains the models of the optimization problem

H) objective_function : contains the objective functions for the optimization problem

I) parameters_vector : contains the functions that create the parameters vector + parameters script that can be modified by user

J) Run_simulation : Contains Files that run the simulations

K) index_script: contains scripts about indexing of the different controllers

L) old: folder that is going to be removed soon

Compile_controllers.m With this file you can basically run every simulations created with the kynematic model of the gokart. You must set the configuration at the beginning of the file:

NUM_Vehicles: 1,2,3 supported

Compiled: yes or no. Yes to skip compilation phase (Controller must be already compiled)

Simulation: yes or no. Yes to run the simulation

TestAlpha1shot: 'yes' or 'no'. yes if you want to test results with different alpha. It requires both IBR and PGalpha compiled (TO BE IMPROVED)

LEPunisher: yes or no. Cost function option: yes if we want to penalize the lateral error only if the gokart is on the left side of the centerline

For 2 vehicles only:

Condition: 'cen' and dec'. 'cen' means centralized controller (One controller for both), dec means decentralized controller (both agents have their own controller)

Game: 'IBR', 'PG'. Iterated Best Response vs Potential Game. IBR is 'dec' only

Alpha: yes or no. yes if you want consider in PG the following cost function: alpha1J1+alpha2J2+Jcol. Only available for 'cen' condition and 'PG'