1
0
mirror of https://github.com/andre-wojtowicz/torcs-client-fuzzy-cpp synced 2024-12-28 18:05:28 +01:00
TORCS C++ client with a driver written in Fuzzy Control Language
Go to file
Andrzej Wójtowicz 8a5032fb1d Update README.md
2016-03-18 15:30:40 +01:00
dep/fl Added project files 2016-03-18 13:04:29 +01:00
src Added project files 2016-03-18 13:04:29 +01:00
.gitignore Initial commit 2016-03-18 12:57:35 +01:00
driver.fcl Added project files 2016-03-18 13:04:29 +01:00
README.md Update README.md 2016-03-18 15:30:40 +01:00
torcs-client-fuzzy-cpp.sln Added project files 2016-03-18 13:04:29 +01:00
torcs-client-fuzzy-cpp.vcxproj Added project files 2016-03-18 13:04:29 +01:00
torcs-client-fuzzy-cpp.vcxproj.filters Added project files 2016-03-18 13:04:29 +01:00

TORCS Fuzzy Client

An exemplary fuzzy-driven AI client for The Open Racing Car Simulator (TORCS). It uses Computational Intelligence in Games client and the project is designed for CIG client-server architecture. The driver is written in Fuzzy Control Language and uses fuzzylite library. The original client files are slightly updated and changed.

The project is released as a proof-of-concept and it ought to be improved as, for example, an assignment for students. The driver uses only few available input variables; more about sensors and effectors of a car and the server can be found in Simulated Car Racing Championship: Competition Software Manual.

To build the project you must provide fuzzylite 5.1 libraries. In the Release section you can find compiled program with FCL driver.

Usage

$ .\torcs-client-fuzzy.exe
Config:
        host: localhost
        port: 3001
        id: SCR_Fuzzy
        max steps: 0
        max episodes: 0
        track name: unknown
        stage: unknown
        fcl file: driver.fcl

Sending init: SCR_Fuzzy(init -90 -75 -60 -45 -30 -20 -15 -10 -5 0 5 10 15 20 30 45 60 75 90)
Driver identified
trackPos: -0.299, angle: -0.000, speedX: 61.793, track9: 200.000, steer: -0.000, accel: 1.000, brake: 0.000
Client shutdown
Bye bye!

Remarks

The project is notably inspired by TORCS Fuzzy Logic Robot, written by my university classmates.