TY - GEN
T1 - A modular parametric architecture for the TORCS racing engine
AU - Onieva, Enrique
AU - Pelta, David A.
AU - Alonso, J.
AU - Milanés, V.
AU - Pérez, J.
PY - 2009
Y1 - 2009
N2 - This paper presents our approach to TORCS Car Racing Competition 2009, it is based on a complete modular architecture capable of driving automatically a car along a track with or without oppents. The architecture is composed of five simple modules being each one responsible for a basic aspect of car driving. The modules control gear shiftings, steer movements and pedals positions by using of simple functions meanwhile the allowed speed in a certain track segment is managed by a simple TSK fuzzy system. Additionally, a module is in charge of modifying outputs' values in case of other cars presence in the environment, with the aim to overtake and avoid collisions with other cars. As a first approach, we provide a "hand-tuned" version of the controllers that allow to achieve very good results, and excellent ones in some particular tracks when compared with last year's competitors. The modules are highly intuitive and these preliminary results open the way to apply soft computing techniques to perform an automatic parameters adjustment for future competitions. Moreover, the modularity of the architecture will allow us to replace or add modules in future, as a way to enhance particular features of the controller.
AB - This paper presents our approach to TORCS Car Racing Competition 2009, it is based on a complete modular architecture capable of driving automatically a car along a track with or without oppents. The architecture is composed of five simple modules being each one responsible for a basic aspect of car driving. The modules control gear shiftings, steer movements and pedals positions by using of simple functions meanwhile the allowed speed in a certain track segment is managed by a simple TSK fuzzy system. Additionally, a module is in charge of modifying outputs' values in case of other cars presence in the environment, with the aim to overtake and avoid collisions with other cars. As a first approach, we provide a "hand-tuned" version of the controllers that allow to achieve very good results, and excellent ones in some particular tracks when compared with last year's competitors. The modules are highly intuitive and these preliminary results open the way to apply soft computing techniques to perform an automatic parameters adjustment for future competitions. Moreover, the modularity of the architecture will allow us to replace or add modules in future, as a way to enhance particular features of the controller.
UR - https://www.scopus.com/pages/publications/71549141992
U2 - 10.1109/CIG.2009.5286466
DO - 10.1109/CIG.2009.5286466
M3 - Conference contribution
AN - SCOPUS:71549141992
SN - 9781424448159
T3 - CIG2009 - 2009 IEEE Symposium on Computational Intelligence and Games
SP - 256
EP - 262
BT - CIG2009 - 2009 IEEE Symposium on Computational Intelligence and Games
T2 - CIG2009 - 2009 IEEE Symposium on Computational Intelligence and Games
Y2 - 7 September 2009 through 10 September 2009
ER -